Lecture Notes in

Mathematics
Edited by A. Dold and B. Eckmann

1023

Stephen McAdam

Asymptotic Prime Divisors

Springer-Verlag
Berlin Heidelberg New York Tokyo 1983
Author
Stephen McAdam
Department of Mathematics, University of Texas at Austin
Austin, Texas 78712, USA

A M S Subject Classifications (1980): 13A17, 13 E05

ISBN 3-540-12722-4 Springer-Verlag Berlin Heidelberg New York Tokyo

ISBN 0-387-12722-4 Springer-Verlag New York Heidelberg Berlin Tokyo

This work is subject to copyright.All rights are reserved,whetherthe whole or part of the material
is concerned,specificallythose of translation,reprinting, re-useof illustrations,broadcasting,
reproduction by photocopyingmachineor similar means,and storage in data banks. Under
§ 54 of the GermanCopyright Law where copies are madefor other than private use, a fee is
payableto "VerwertungsgesellschaftWort", Munich.
© by Springer-VerlagBerlin Heidelberg 1983
Printed in Germany
Printing and binding: Beltz Offsetdruck, Hemsbach/Bergstr.
2146/3140-543210
TO MARTHA
 ACKNOWLEDGMENTS

Numerous people have participated in the study of asymptotic prime divisors,

and I have tried to acknowledge, in the text, a sampling of their contributions.

To do so entirely would be impossible, and I hope I have been fair in my selection.

Certain people have been particularly helpful to me, as much in stimulating conver-

sations as in specific results. I offer my gratitude to Paul Eakin, Ray Heitmann,

Dan Katz and Keith Whittington. My special thanks goes, as it does so often, to

Jack Ratliff.

Part of my research was supported by the National Science Foundation, for which

I am grateful.

Nita Goldrick typed the manuscript. Her great skill and patience eased a

difficult task.
 TABLE OF CONTENTS
Page

INTRODUCTION VIII

CHAPTER I : A (I) and B (I) 1

CHAPTER II: A (I)-B (1) 8

CHAPTER III : A (I) 12

C H A P T E R IV : A Characterization of A (I) 26

C H A P T E R V: Asymptotic Sequences 32

CHAPTER VI : Asymptotic Sequences O v e r Ideals 42

CHAPTER VII: Asymptotic Grade 55

. --w
CHAPTER VIII : When A = A 61

CHAPTER IX : Conforming Relations 68

CHAPTER X : Ideal Transforms 76

CHAPTER XI : Miscellaneous 89

APPENDIX: Chain Conditions 1IO

REFERENCES 113

LIST OF NOTATION ll6

INDEX Il7
 INTRODUCTION

Asymptotic prime divisors represent the interface of two major ideas in the

study of co~utative Noetherian rings. The first, the concept of prime divisors,

is one of the most valued tools in the researcher's arsenal. The second is the

fact that in a Noetherian ring, large powers of an ideal are well behaved, as shown

by the Artin-Rees Lerm~a or the Hilbert polynomial.

Although its roots go back further, the recent interest in asymptotic prime

divisors began with a question of Ratliff: What happens to Ass(R/I n ) as n gets

large? He was able to answer a related question, showing that if [ is the integral

closure of I, then Ass(R/I n ) stabilizes for large n. In a later work, he also

showed that Ass(R/I n ) ~Ass(R/In+l). (Earlier, Rees had shown that if

P c Ass(R/In), some n, then P e Ass(R/l TM) for infinitely many m.) Meanwhile,

Brodmann answered the original question, proving that ~ss (R/In) also stabilizes

for large n. Since then, the topic of asymptotic prime divisors has been growing

rapidly, the latest development being the advent of asymptotic sequences, a useful

and interesting analogue of R-sequences.

These notes attempt to present the bulk of the present knowledge of asymptotic

prime divisors in a reasonably efficient way, to ease the task of those wishing to

learn of, or contribute to the subject. Modulo some gnashing of teeth, and rending

of garments, it was both educational and satisfying to write them. I hope that

reading them is the same.

The first chapter shows that for an ideal I in a Noetherian ring R,

Ass(R/I n ) stabilizes for large n, as does Ass (In'l/In), the respective stable

values of these two sequences are being denoted A (I) and B (I). Also B (I) is

characterized as the contraction to R of prime divisors Q of t-l~ with

It ~ Q, where ~=~[t-l, lt] is the Rees ring of R with respect to I.

Chapter Two shows that A (I) - B (I) ~ A s s R, and that P e A (I) - B (I) if

and only if there is a k ~ i such that p(k) is part of a primary decomposition

of In for all sufficiently large n.

 VIII

Chapter Three shows that Ass(R/~) c_ Ass(R/I 2) c_,.. , and that this sequence
--,k -~¢ *
eventually stabilizes to a set denoted A (I). Furthermore, A (I) c_ A (I). It

also developes several technical results useful for dealing with A (I), the most

important of these being that in a local ring s P e A (I) if and only if there are

primes q • c p* in the completion R* such that q* is minimal, p* NR =P and

p • /q* e ~ (IR* + q*lq*)

In Chapter Four, it is shown that if R is locally quasi-unmixed, then

P e A (I) if and only if height P = g ( ~ ) , the analytic spread of ~ . Since a

complete local domain is locally quasi-unmixed, this result meshes nicely with the

one mentioned from Chapter Three.

Chapter Five introduces asymptotic sequences: A sequence x I , ..., xn such

that (Xl,..., X n ) ~ R and for i = 0, ...,n-l, xi+ 1 4 U [ P e A ((Xl,..., xi))]. In

a local ring (R,M) it is shown that x I , ..., x n is an asymptotic sequence if and

only if height((xl,..., Xn)R + q /q ) = n for each minimal prime q of the com-

pletion. This is then used to show that for a given ideal I in any Noetherian

ring, all asymptotic sequences maximal with respect to coming from I have the

same length, denoted gr I. It is then shown that asymptotic sequences are to

locally quasi-unmixed rings as R-sequences are to Cohen-Macaulay rings.

In Chapter Six, the sequence x I , ..., x n is called an asymptotic sequence over

the ideal I if (l,x I,..., Xn) @ R and for i =O,...,n-l,

xi+ 1 ~ U [P e A ((l,x I,..., xi))}. It is shown that in a local ring, all maximal

asymptotic sequences over I have the same length.

Chapter Seven proves that in a local ring, the grade of R/I n stabilizes for

large n, and gives partial results concerning gr(R/In).

Chapter Eight identifies, with one possible exception, all Noetherian rings for

which A (I)=A (I) for all ideals I.

In Chapter Nine, asymptotic prime divisors play a minor role in proving the

following unexpected result. Let P be prime in a Noetherian domain. Then there

is a chain of ideals p = l 0 c II c . . . c In with the following property: Let Q

be a prime containing P, and let j be the largest subscript such that lj ~ Q.

Then p c Q satisfies going down if and only if j is even.

 IX

In Chapter Ten, we consider a local ring ~,M) and the ideal transform of

M, T(M). Previously it was known that the following two statements are equivalent:

(a) T(M) is an infinite R-module (b) The completion of R contains a depth i

prime divisor of zero. Our main result adds two more equivalent conditions:

(c) M e A*(J) for every regular ideal J (d) There is a regular element x with

M e A * (J) for all J~xR. Here J~l if for some n and m, i n and jm have

the same integral closure. Motivated by statement (d), we then discuss the possi-

bility of defining a strong asymptotic sequence x I ,..., Xn with (xI , ..., X n ) ~ R

and for i=0,...,n-I xi+ I @ U [ P e n A * ( J ) I J ~ ( X l , . . . , xi)}, in the hope that such

a sequence will stand in relation to prime divisors of zero, as asymptotic sequences

stand to minimal primes. This program is carried out for n = i and 2.

Chapter Eleven is aptly titled Miscellaneous. It contains topics (of varying

worth) which did not fit elsewhere.

The study of asymptotic prime divisors frequently impinges on that of the

structure of the spectrum of a Noetherian ring, often referred to as the study of

chain conditions. I have tried to keep to a minimum the amount of knowledge of

chain conditions necessary to read these notes. In the Appendix, I list those

definitions and basic results (with references for the curious reader) which are

referred to in the text.

 CHAPTER I: A (I) and B (I)

DEFINITION. Let I be an ideal in a Noetherian ring R. For n = 1,2,3,..., let

A (I, n) = Ass (R/I n ) and let B(l,n) =Ass(In-I/In).

In [R3], Ratliff asked about the behavior of the sequence A(l,n) (and

showed that a related sequence stabilized, see Chapter 3). In [BI], Brodmann

showed that both sequences A(l,n) and B(l,n) stabilize for large n, as we now

show. Recall that the graded Noetherian ring T = ER n , n ~ 0 is homogeneous if

T = Ro[RI]. Our first lemma is well known.

LEMMA I.I. a) Let En> 0 R be a Noetherian homogeneous graded ring. Then there
n
is an ~ such that for n >__ ~, (0 :RI) N R n = 0 .

b) Let I be an ideal in a Noetherian ring. Then there is an ~ such that for

n ~ ~, (In+l : I) n I ~ = I n.

Proof: a) Let (0 :RI) be generated by homogeneous elements al,... , as " Let

= l+max[deg ai}. If x=Er.a.l i e (0 : RI) N R n with n ~ ~, then we may assume

the r.l are homogeneous and have r i e RIT. Thus riai = 0, and so x = 0.

b) Let ER = E l n / l n+l and pick ~ as above. Say n > ~ and let

n
x e (in+l : I) N i ~" Suppose that x ~ In . Let x e Ik - Ik+l ' and note that

~ k < n. Since xl ~ I n+l ~ I k+2, with ~ e Ik/l k + l = R k we have

0#~ e (0 :R I) N R ~ , contradicting part a.

LEMMA 1.2. Let T = ER be a Noetherian graded ring. Let I be a homogeneous

n
ideal and let c be a homogeneous element. Suppose that S is a multiplicatively

closed subset of R0 and that (I : c) N S = @. Then there is a homogeneous element

d, such that (I : cd) is prime and (I : c d ) N S = ~ .

Proof: Among all homogeneous d' with (I : cd') N S = @, choose d so that

(I : cd) is maximal. It is enough to take homogeneous x and y not in (I : ed)

and show xy ~ (I : cd). Suppose, contrarily, that xy e (I : cd). Then

x e (I : cdy) so that (I : cdy) is strictly larger than (I : cd). Thus there is

an s c S N (I : cdy). Now y e (I : cds), showing that this ideal is strictly

larger than (I : cd). Thus there is an s' e S N (I : cds). This gives

ss' e s N (I : cd), a contradiction.

PROPOSITION 1.3. [ME] Let T = ~n>0 Rn be a Noetherian homogeneous graded ring.

Then there exists an m such that ASSR0 ~m ) = ASSR0 (Rn) for a l l n ~ m.

Proof: Let P e U ASSR0 (Rk), k = 0,1,2, . . . . Then for some homogeneous c e T,

P = (0 : C)Ro . Clearly P = (0 : c) T ~ R 0 and by Lermna 1.2, for some homogeneous

d e T we have P * = (0 : cd) prime in T and P * N R 0 =P. As ASST(T ) is finite,

we see that U ASSR0 (R~) is finite•

Now select ~ as in Lemma i.I and say n > ~° If P c ASSR0 (Rn) write

P = (0 : C)R0, c e Rn . As n ~ ~, P = (0 : cRI)R0 ° Since cR I ~ Rn+ I , we have

P c ASSR0 (Rn+l). Thus ASSR0 (Rn) ~ ASSR0 (Rn+l) for n > ~° As we already

have U ASSR0 (Rk) finite, the result follows.

COROLLARY 1.4. (Brodmann [BI]) Let I be an ideal in the Noetherian ring R.

The sequence B(l,n) stabilizes•

Proof: Apply the proposition to Eln-I/In°

COROLLARY 1.5. (Brodmann [BI]) Let I be an ideal in the Noetherian ring R.

The sequence A(l,n) stabilizes.

Proof: The exact sequence 0 + In/l n+l ~ R/I n+l + R/I n + 0 shows that

A(l,n+l) ~ A(l,n) UB(I,n+I). For large n, we already have B(l,n+l) = B(l,n)

A(I,n). Thus A(I,n+l) ~ A(I,n), and the result is clear since A(I,n) is finite.

Note that for an ideal I in a Noetherian ring R, B(l,n) ~ A(l,n)° The

following example, due to A. Sathaye, shows that neither sequence is monotone.

EXAMPLE• Let k be a field and n a positive integer, Let R = k [ x , z I ,..., Z2n]

with the restrictions that 2i-I z2i

xz2i_l= 2i for i = 1,2, . ..,n, and z~z i = 0 for
I ~ i,j ~ 2n. Let I = (z I , z 2 ,..., Z2n) ~ P = (x,z I ,,.., Z2n). Then for

I < i < 2n, P e B(l,i) if i is even, while P 4 A(l,i) if i is odd.

2i-I @ 12i 2i-i _ 12i,

Proof: Since z2i_l and Pz2i_l c we have P e B(l,2i) for

i < i < n. To see that P @ A(I,s) for s odd, 1 < s < 2n, note that

P 4 A(I,I) since I is prime. Now for i ~ q ~ 2n, the residues of the set
u
Tq = [z } U [z u2
2 ...z 2 ~n lu2+,,,+U2n = q, 0 ~ u l. < i} form a generating set for

lq/l q+l over k[x]. If q is even, there are no relations, and T gives a
q
free basis •
If q is odd, there is the unique relation xz qq e I q+l . Suppose

P e A(I,s). Then for some w ~ I s , Pw ~ I s, Consider r such that w e I r - I r+l.

By the previous remarks, xw e ir+l shows that r is odd° Furthermore, it can

be seen that xw 4 Ir+2. Thus s= r+ i and so s is even.

DEFINITION. For I an ideal in a Noetherian ring, the eventual constant values

of the sequences A(I,n) and B(l,n) will be denoted A (I) and B (I),

respectively.

The fact that A (I) and B (I) behave well under localization is straight-

forward, and yet we will use it so often that we state it formally.

LEMMA 1.6. Let I c p be ideals in a Noetherian ring, with P prime. Then

P e A (I) (respectively P e B (I)) if and only if PS e A (Is) (respectively

PS e B ( I s ) ) , f o r any m u l t i p l i c a t i v e l y closed set S disjoint from P.

The next result will lead to some interesting applications of asymptotic

prime divisors. As this result will be used again when discussing the integral

closure of an ideal (Chapter 3), we give it here in full generality,

If J is an ideal of R, we will use ~ to denote the integral closure of

J. Thus ~ = {x e RIx satisfies a polynomial of form X n + jl Xn-l+,..+ Jn = 0, with

Ji e ji}. Recall that R is the integral closure of R.

PROPOSITION 1.7. [M3] Let P be a prime ideal in a Noetherian domain R. There

is an integer n ~ i with the following property: If I is an ideal of R with

I ~ pn, and if there exists an integral extension domain T of R and a

Q e spec T with QN R = P and Q minimal over IT, then P e Ass(R/I).

Proof: Let PI'''" Pm be all of the primes of R which lie over P. Select

u i e Pi'U Pj • j ~ i, and let S = R [ u I,..., Urn]. Notice that Pi is the unique

prime of R lying over pi=PiNS. Let (Vi, Ni) be a D.V.R. overring of S

with N i N S = Pi " Since S is a finitely generated R-module, we can choose b e R

with bS c_ R. Pick n sufficiently large that b 4 N n, i= 1,2,...,m.

Suppose that I c pn and that T is an integral extension domain of R

containing a prime Q with QNR =P and Q minimal over IT. We first reduce

to the case that T = S. Clearly we may assume T = ~, and by going down we may

replace ~ by R. Finally since P. is the only prime of R lying over Pi'

i

i = 1,2,°..,m, by going up we replace R by S.

We now have T=S, and of course Q = Pi for some i = 1,2,...,m. We localize

making P maximal in R. Since Pi is minimal over IS, there is an integer

k ~ I and an s e S -pi with k c. IT.

sPi Using
. bS. c R, we
. have bsP k c bsp~ c

bit ~ I. Furthermore, we claim bs 4 5. If bs e [ ~ pn ~ p~ ~ N ~ = N ni, then

since s 4 Pi implies s is a unit of Vi, we have b e N ni' contradicting our

choice of n. Thus bs 4 1 but bsP k ~ I, showing that pk consists of zero

divisors modulo I. As P was maximal, P c Ass(R/I).

COROLLARY 1.8. Let I be an ideal in a Noetherian domain R and let T be an

integral extension domain of R. If Q is prime in T and minimal over IT,

then QNR e A* (I).

Proof: Let P = Q n R, and choose n as in the proposition. Then P c A(l,m)

for m > n.

The following fact about the integral closure of a Noetherian domain appears

to depend upon knowledge of asymptotic prime divisors.

PROPOSITION 1.9. Let R be a Noetherian domain. Let J = (b I ,..., bm) be a

finitely generated ideal of the integral closure R. Then the number of primes of

minimal over J is finite.

Proof: Let S = R [ b I ,..., bm] and let I = (b I ,..., bm)S. Thus S is Noetherian

and I R = J. If Q e spec R and Q is minimal over J, then by Corollary 1.8,

Q ~ S e A (I). Since A (I) is finite and since only finitely many primes of

lie over a given prime on S, we are done.

We generalize [N, 33.11].

PROPOSITION i. I0° Let R c T be an integral extension of domains with R

Noetherian. Let Q be a height n prime of T and let P = Q n R. Then

grade P ~ n. If grade P = n, then for any R-sequence a I ,°.., an coming from

P, P is a prime divisor of (aI ,.°., an).

Proof: We induct on n. For n = i, pick a # 0 in P. Since height Q = I, Q

is minimal over aT. By Corollary 1.8, for sufficiently large k, P is a prime

divisor of akR. It is not difficult to now see that P is also a prime divisor

of aR.

For n > I, suppose grade P ~ n and let a I ,..., an be an R-sequence

coming from P. We claim height(a I ,..°, a n ) T = n . If not, say q e spec T,

height q < n and (aI, ..., an)T ~ q. By induction, grade q n R ~ height q < n,

contradicting that a I ,..., an is an R-sequence in q A R. Thus the claim is true,

and so Q is minimal over (al,..., an)T. By Corollary 1.8, for large k we

k
have P a prime divisor of (aI ,..., an ) in R. As al ,,.., an is an R-sequence,

P is also a prime divisor of (a I,..., an) by [KI, Section 3-1, Exercise 13].

In Chapter 5 we strengthen Proposition I.I0, replacing "height Q" by

"little height Q".

The next three propositions give easy circumstances under which a prime must

be in A (I).
PROPOSITION I°ii. Let I be an ideal in a Noetherian ring, and let the prime P

be minimal over !. Then P e A (I). Also P e B (I) if and only if height P > 0.

Proof: Since P is minimal over In for all n, P e A(l,n), and so P e A*(1).

For the second statement, localize at P, so that I is P-primary. Now

height P > 0 if and only if I is not nilpotent. If I is nilpotent, clearly

P @ B * (I). If I is not nilpotent, then for all n in/in+l is a nonzero module

(by Nakayama's Le~mna) which must have at least one prime divisor. However P is

the only possibility. Thus P e B(l,n) for all n.

PROPOSITION 1.12. Let I c p with P a prime divisor of 0 in a Noetherian

ring. Then P e A (I).

Proof: Localize at P and then write P = (0 : c). For n large enough that

c ~ In , clearly P = (In : c).

Our next proposition generalizes Proposition I.ii° The lemma is due to

Ratliff.

LEMMA 1.13. Let Q c p be primes of the Noetherian ring R such that Q is a

prime divisor of O° Then there is an integer n > 0 such that for any ideal J

of R with J c pn and P minimal over Q+J, we have P e Ass(R/J).

Proof: Localize at P. Let qlN'''N qr be a primary decomposition of 0 with

ql primary to Q. Choose 0#x e q2N"'Nqr' and pick n such that x ~ pn

Suppose that p e Ass(R/J) and p # P. Since P is minimal over Q + J, we have

Q ~_ p~ Thus in Rp, 0 = (q2)pN...n (qr) p so that X R p = 0. This shows that x

is in every p-primary ideal. However, j ~ pn shows that x 4 J. Thus

P e Ass(R/J), using primary decomposition.

PROPOSITION 1.14. Let I,P,Q be ideals in a Noetherian ring with Q a prime

divisor of 0, and P a prime minimal over Q + I. Then P e A (I).

Proof: With n as in Lemma 1.13, P e A(l,m) for all m >_ n.

 Later (Proposition 2.5) we will strengthen Proposition 1.14 to say that if in

addition P # Q, then P ~ B (I).

We give a characterization of B (I) in terms of the Rees ring of R with

respect to I, that is, the ring ~= R[t-l, It] with t an i n d e t e r m i n a t e .

PROPOSITION 1.15. Let I be an ideal in the Noetherian ring R, and let

=R[ t-l, It] be the Rees ring of R with respect to I. Then P c B * (I) if and

only if there is a prime divisor Q of t-l~ such that It ~ Q and QNR=P.

Proof: Let P e B (I). Consider ~ as in Lemma l.lb, and choose n > ~ with

P e Ass(In/I n+l ). Write P = (In+l : c) with c e In . Since ctn e ~R, note that

(t-l~: ctn) N R = (In+l: c ) = P . By eemma 1.2 there is a atm e ~ such that

Q = (t-l~ :dct n-Mn) is prime in ~ and QNR=P. We must show that It ~ Q° Since

Q is a proper ideal, dctn+m= (drm)(ct n) @ t-l~. Thus m > 0 and dc @ In+m+l,

By Lemma I.i, (In+m+2 : I) N I ~ = I n+m+l, and since c e In we must have

de ~ (In+m+2 : I) o Therefore It ~ (t'l~ : dct n+m) = Q as desired.

Conversely, suppose that Q = (t'l~ : gtk) with g c Ik, that QNR=P, and

that It ~ Q o Pick ht e I t - Q . Clearly Q = (t'l~ : ghmt k+m) for all m > O.

Thus P = (Ik+n~l : ghm). Since ghTM e Ik+m and m is arbitrary, we have

P e B (I).

We close the chapter with a question. We have seen that the sequence

A(I,I),A(I,2),... is not increasing. Ratliff asks whether A(I,I) N A (I),

A(I,2) N A (I),... is increasing?

 CHAPTER II: A (I)-B (I)
*
In this chapter, we study primes contained in A (I) but not B* (I), our

main result being that such primes must be prime divisors of zero.

L E M M A 2.1. Let I be an ideal in a Noetherian ring R and (by L e m m a i.i) suppose

for n > ~ we have (In : I) N I ~ = I n-l. If P is prime in R and if P = (In : e)

with n > ~ and c e i~ ' then P e B * (I) o

Proof: Since cl c cP ~_ I n , we have c e (In : I) N I Z = I n-l. For j ~ 0, clearly

p c (I n+j : cIJ). Conversly, if r e (In+j : cI j) then rcl j-I ~ (In+j : I) N I ~ =

I n+j'l , so that r e (In+j-I : elJ'l). Iterating, we find r e (In : c) = P . Thus

P= (In+j : cI j) for j = 1,2,... . Now we already have c e In-1 , so cIJ~_I n+j-l.

Thus P e B(l,n+j) for j=l,2,....

PROPOSITION 2.2 [ME] Let ! be an ideal in a Noetherian ring R. If P e A (I)-

B (I) then P is a prime divisor of zero.

Proof: We may localize at P. Since P e A (I), for all large n we have an

xn e R with P= (In : Xn), and by Lemma 2.1 we have Xn @ I~" To show that P

is a prime divisor of zero, it is sufficient to show this in the case that (R,P)

is complete, w h i c h we now assume. Let V = (I~ : P)/I ~ and for n > ~ let

V = [(I n : P ) + I ~ ] / I ~. Now PV=0, so V is a finite dimensional vector space

n
over R/P. Clearly x taken modulo I~ is a nonzero element in the subspace
ii
V Since V n + I ~ V n , we see that f~ V n @ O, by finite dimensionality. Let
n
be in this intersection, and let h e (I ~ :P) - I ~ be a preimage. Since

%eV n write •= d
n
+i
n
with d
n
e (In : P) and i e I b.
n
For m > n
- -
we have

- ' - - i e (I n : I) N I ~ in-I (_- p n - l ,

I(i n im) ~_ p(i n im) = p ( d n - din) c~ I n . Thus in m = --

showing that the sequences [in} and [d n] are Cauchy sequences. Let in + i

and d n -~ d. Since i e I~ and % @ I ~, d # 0. Finally, since dn P c_ I n c_ pn,

dP ~ N p n = O, concluding the proof.

PROPOSITION 2.3. Let I be an ideal in a Noetherian ring R. Then P e A (I) -

B (I) if and only if there is an integer k > I such that for all sufficiently

large n, the k-th s y m b o l i c power p(k) is part of a primary decomposition of In •

Proof: We may assume R is local at P. In this context, p(k) = p k . First

suppose that for all large n, that In has primary decomposition

in = P k N qnl N...N qnm" Clearly P e A*(1). Let n ~ k+l and let P = (In : c).

To show P @ B (I), we desire c ~ I n'l. Since qni is primary~to a prime prop-

erly contained in P, Pc ~ I n shows that c e qnl N...N qnm ° Since P= (In : c)

clearly c ~ In . Thus c @ pk However in-I ~ pk, so c @ I n'l as desired.

Conversely, suppose that P e A (I) - B (I). Let Z be as in Lemma i.I.

Since ~ can be increased, also assume that A(I,~) = A ( I , ~ + I ) =...= A (I).

Suppose A (I) = {P'QI ''''' % } and for n ~ ~ let the primary decomposition of

In be q n N qnl N...N qnm with RAD qn = P" We may assume that qn+li ~ qni ' so

that if Jn=qnl N...n qnm' then Jn+l ~ Jn " Claim J n N I ~ = In . One inclusion

is obvious. Consider c e J N I ~, and by way of contradiction, assume c @ In .

n
Thus (I n : c) is proper, and so is contained in P° Now for some large t,

pt ~ qn " Thus ptc ~ qn c ~ q n N Jn = In" Therefore pt ~ (in : c) ~ P. Using

Lemma 1.2, we see that for some r e R, (In : rc) is a proper prime ideal, and

since pt ~ (in : c) ~ (In :rc), we must have (In : rc) = P . Now c e I ~, so

rc e i ~ ' and by Lermma 2.1 we have P e B * (I). This is a contradiction, proving

the claim that Jn N I ~ = In, Now since Jn+l c-- Jn ' we have Jn c- J~ " Thus

I n = J n N I ~ = J n N J ~ N q ~ = J n N q~ . Finally, take k large enough that pk c_ q~ .

Then for n >_ k, I n = J n N pk, giving the desired primary decomposition.

Consider an ideal I in a Noetherian ring R. Since (0 : I) ~ (0 : 12) ~ .... ,

there is a k with (0 : Ik) = (0 : Ik+l) =... . Call this ideal J. The following

lemma, due to P. Eakin, shows that reducing modulo J can be useful.

LEMM 2.4. Consider I and J = (0 : Ik) as above. Let I' = I modulo J. Then

A*(I') =B*(I'). Also P e B*(1) if and only if J ~ P and P' e A*(I').

 10

Proof: Suppose P' e A*(I') -B*(I'). Then by Proposition 2.2, P' = (0 : c') for

some c' e R'. Since I' c p', we have c'l '= 0. Thus cl c j = (0 : Ik), so that

clk+l= 0. Therefore c e (0 : Ik+l) = (0 : Ik) = J, and so c ' = O. This contradicts

that (0 : c') = P ' # R ' , and shows that A (I') = B (I').

Now consider ~ as in Lemma I.I. We easily see that (In+k : Ik) N I ~ = I n

for large n. Suppose that x e J N I ~. Then xl k = 0 ~ I n+k for all big n, so

that x e (In+k : I k) n I ~ = I n for all big n. That is, j n i ~ c Q i n" n = 1,2, ....

We next claim that for large n, In/l n+l m l'n/l 'n+l. To see this, map

a + l n+l to a' + I 'n+l and note that the map is injective since In N J c _ In+l '

from above.

Suppose P e B (I). This is still true upon localizing at P, and so ~ is

not n i l p o t e n t . Thus J ~ P, or else we w o u l d h a v e I ~ = O. By t h e a b o v e i s o m o r p -

ism, P is a prime divisor on l'n/l 'n+l, for large n. However, J annihilates

this m o d u l e , which therefore is an R ' - m o d u l e having P' as a prime divisor. Thus

P' e B (I'). The converse is similar.

We use this lermma to strengthen Proposition 1.14.

PROPOSITION 2.5. Let I ,P ,Q be ideals in a Noetherian ring R, with Q a

prime divisor of zero and with P a prime minimal over Q + I. If P ~ Q, then

P ~ B (I).

Proof: Let J = (0 : Ik) as in Lemma 2.4. Since P#Q, clearly I ~_ Q, and so

j c_ Q. Writing Q= (0 : c) for some c e R, we have Q c (j : c). We claim

equality. If x e (J : c) then since I ~_ Q, pick y e Ik - Q . As

xc e J = (0 : Ik), we have xcy=0. That is, xy e (0 : c ) = Q . Since y ~ Q, x e Q

as claimed.

We now work modulo J, using primes to denote images. We have just seen

that Q' is a prime divisor or zero in R'. As P' is minimal over Q'+I',

Proposition 1.14 gives P' e A*(I'). By Le_nnna 2.4, P e B (I).

 11

Combining Propositions 2.2 and 1.12, we see that A (1)=B (1) U[P e Ass(R) I

I ~P}. We know very little of what can be said about the overlap, B (I) N [ p

Ass(R) II ~ P } . We g i v e an example i n which I c P1 c P2 c P3 with P1 ' P 2 ' P3

all in Ass~), and PI and P3 are in B (I), but P2 ~ B (I).

EXAMPLE. Let T=K[X,Y,Z,W]. Let Pl = (X,Y), P2 = (X,Y,Z), and P3 = (X,Y,Z,W),

while ql = (X), q2 = (Z), and q3 = (X,Z,W). Let L=plP2P3qlq2q3 , let R = T/L,

let PI' P2' P3' QI" Q2 ' Q3 he (respectively) the images of PI' P2' P3' ql ' q2' q3

and let I be the image of (Y). Now I ~ P1 ~ P2 ~ P3 and these three primes

are in Ass(R). Also Q1 and Q3 are in Ass(R). Since P1 =QI + I and

P3 = Q3 + I, by Proposition 2.5 we have P1 and P3 in B (I). We claim that

P2 ~ B (I)° Invoking Lermma 2.4, we note that y3 e plP2P3 , so for k ~ 3 we

have (0 : Ik) =QIQ2Q 3 By Lemma 2.4, we must show that p2/qlq2q3~A*((Y)+qlq2q31

However the image of Y modulo qlq2q3 is a regular element, and so we need only

show that p2/qlq2q3 is not a prime divisor of (y) + qlq2q3/qlq2q3 , which is clear.

It is natural to ask whether in Proposition 2.5, the condition P minimal over

Q+ I can b e weakened t o j u s t P e B (Q+ I ) . The answer i s no. I n t h e above e x -

ample, we a l r e a d y have P2 ~ B ( I ) . However, s i n c e P2 =Q2 +(Q1 + I ) , Proposition

2.5 shows t h a t P2 e B (Q1 + I ) "

 CHAPTER III: A (!)

In this chapter, we study another sort of asymptotic prime divisor. Recall

that for an ideal I, ~ denotes the integral closure of I. We will let 2(I,n) =

Ass(R/In). Our main goals will be to prove that ~(I,I) ~ A(I,2) ~... , that this

sequence eventually stabilizes at a set denoted A (I), and that A (I) ~ A (I).

These results were first proved by Ratliff in [R3] and [R8]. (In [Rs2], Rees out-

lines a different approach to these ideas.) The essence of the arguments needed are

mostly easily seen when R is a domain. We treat that case first, following Rat-

liff's trail.

LEMMA 3.1. Let R be a Noetherian domain with integral closure R. Let (V,N) be

a D.V.R. overring of R and suppose that V is a localization of an integral ex-

tension of a finitely generated extension of R. If the transcendence degree of

V/N over R/NNR is 0, then height(MAR)= io

Proof: Let V be a localization of an integral extension of A, with A finitely

generated over R. Now A c ~ c V, and A is an integral extension of a finitely

generated extension of R. Since the transcendence degree of A/NN ~ over R/NN

is 0, and also since h e i g h t ( N N A ) =height N = i, we see that NNA is isolated

among primes of ~ lying over N N R. By the Peskine-Evans formulation of Zariski's

Main Theorem [E], ~N~ = ~NQ~" Thus height(MAR) =height(NNA) = I.

LEMMA 3.2. Let a# 0 in the Noetherian domain R. Let P be a prime divisor of

anR for some n > 0. Then in R there is a height 1 prime divisor p of aR

with pNR=P.

Proof: Write P = (anR : c). Since anR = anRn R, P = (anR : c)~N R. Now the Krull

domain ~ has A.C.C. on ideals of the form (b : d). Thus the argument used in

proving Len~a 1.2 shows that (anR: c)~ can be enlarged to a prime divisor p of

anR, with p N R = P. Since R is a Krull domain, height p = I.

 13

LEMMA 3.3. Let I be an ideal in a Noetherian domain Ro Let i = R [ t ' l , lt] be

the Rees ring of R with respect to I. If Q is a prime divisor of t'~R for

some n > 0, then either It ~ Q or Q Q R = 00

Note: Since I= t ' I ~ N R ~ Q N R , if I# 0 we must have It ~ Q .

Proof: Using Lemma 3.2 we see that ~ contains a height 1 prime q lying over

Qo Now (V,N) = (~q, qq) is a D.V.R. Suppose that It ~ Q . Then R/QNR =~/Q

and clearly the transcendence degree of V/N over R[t-I]/QNR[t-II is 0. By

Lemma 3.1 we have height N N R [ t "I] =height Q N R [ t -I] = I. Moreover Q N R [ t "I]

obviously equals (QNR, t-I)R[t'I], since t -I e Q. However, for T a Noetherian

domain and X an indeterminate, only one height 1 prime of ~[X] contains X,

namely X~[X], and it intersects T[X~ at XT[X]. With T=R and X = t -1, we

see that (QNR, t-I)R[t -I] must be just t-IR[t-II, so that QNR= O.

PROPOSITION 3.4. Let I# 0 be an ideal in a Noetherian domain R. Then

A(I,I) ~ ~(I,2) ~ .... This sequence stabilizes to a set denoted A (I). Further-
-4¢ *
more A (I)~ B (I).

Proof: Let P e A(I,n) and write P = (In : C)R for c e R. Then since

In = t'n~N R, P = (t-n~: e)~N R. By Lemma 3.3, no prime divisor of (t'n~ : c)~R can

contain It. Select dt e It with dt a nonzero divisor modulo (t-n~ : c)~ •

Thus (t-n~R : c)~= (t-n~ : cdt)~ and this ideal meets R at P. Since the degree 1

component of t-n~ is (In+l N I)t, we see that P = (In+l N I : cd) = (In+l :cd)

using that d e I. Therefore P e ~(I,n+l), which shows the sequence is increasing.

We will now show that A(I,n) ~ B (I). As B (I) is finite, the rest of the

theorem follows immediately. As above, write P = (t'n~ : c)~N R° As in Lemma 1.2,

enlarge (t-n~ : c)~ to a prime divisor Q of t-n~ with QNR=P. By Lermna 3.3,

It ~ Q. Also Lemma 3.2 gives a height i prime q of ~ lying over Q° By

Proposition I.I0, height q = I implies grade Q = i. Thus Q is a prime divisor

of t'l~° Since It ~ Q, Proposition 1.15 tells us that p=QNR e B (I).

We apply Proposition 1.7 to this context, and easily see the following.
 14

PROPOSITION 3.5. Let R c T be an integral extension of domains, with R

Noetherian. Let I be an ideal of R. If Q e spec T with Q minimal over It,

then QAR e A (I).

We now drop the assumption that R is a domain.

LE~RIA 3.6. If I c J are ideals, then I reduces J if and only if for each

minimal prime q, l + q / q reduces J +q/q.

Proof: One direction is trivial. Thus suppose that ql ''''' qm are all the

minimal primes, and that for each i, I mod qi reduces J mod qi " Then for
jk+l
sufficiently large k we have jk+l ~- i j k + q i' i= l,...,m. Since IJk

in fact we have j k + l = i j k + (jk+IAqi). If n is such that (ql •'" qm )n = O,

then jnm(k+l) = ~ (ijk + (jk+l n qi ))n c IJ nm(k+l)'l ~ j n m ( k + l ) Thus I reduces J.

i=l

LEMMA 3.7. Let P be a prime divisor of ~ for some ideal I in a Noetherian

ring. Then there is a minimal prime q contained in P such that P/q is a

prime divisor of (l+q/q).

Proof: We may assume that R is local at P. Let P = (5: c). Since c ~ ~, I

does not reduce (I,c). Thus Lenlna 3.6 shows that for some minimal prime q,

I+q/q does not reduce (I,c) +q/q. Therefore c+q @ (I+q/q). We easily see

that (P/q)(c+q) c ~ + q/q ~ (I+q/q), and the result follows.

LEMMA 3.8• Let ~ be the Rees ring of R with respect to an ideal I. If Q

is a prime divisor of t-n~ for some n > O, then either It ~_ Q or QN R is

a minimal prime in R.

Proof: By Lermna 3.7, there is a minimal prime q of ~ with Q/q a prime

divisor of (t'n~+q/q). We claim that q N R [ t -I] is minimal in R[t-l]. For

this, let S = {l,t'l,t-2,...}. Since t -I is regular in ~, q N s = ~ . Now

R[t-l] S = ~ s = R [ t ' l , t ] . Since qs is minimal, the claim is obvious. Furthermore,

if p=qnR, we see that p is minimal in R, that qs=PR[t'l,t], and that

the degree n component of q is (InNp)t n (with I n = R if n <__ 0). Since

 15

I n / I n N p ~ In+p/p, we see that ~/q ~ (R/p)[t "l,(l+p/p)t] =~', the Rees ring of

R/p with respect to I + p/p. This isomorphism takes Q/q to a prime divisor

Q' of t'n~ '. By Lermma 3.3, either (l+p/p)t ~ Q' or Q' n (R/p) =0. If

(l+p/p)t ~ Q' then clearly It ~ Q. If Q'NR/p=0, then Q/qnR/p=0, so that

QAR=p is a minimal prime in R.

PROPOSITION 3.9. Let I be an ideal in a Noetherian ring R. Then A(I,I)

A(I, 2) ~ ... and this sequence eventually stabilizes.

Proof: If P is a minimal prime of R and I ~ P, then clearly P is in every

term of our sequence. If P e A(l,n) and P is not minimal in R, then the

proof that P e ~(l,n+l) is identical to the first paragraph of the proof of Prop-

osition 3.4 (using Lemma 3.8 instead of Lermua 3.3). Thus ~(I,I) ~ ( 1 , 2 ) ~ ....

Now suppose P e A(l,n). By Lemma 3.7, there is a minimal prime q of R

with q ~P and P/q is in ~(l+q/q,n). Thus P/q e ~ ( l + q / q ) by Proposition 3.4.

As R has only finitely many minimal primes q, and as A (l+q/q) is finite, we

see that UA(l,n) is finite. Thus A(I,I) ~ A(I,2) ~... eventually stabilizes.

DEFINITION. A (I) will denote the limit set of the above sequence.

Note: We will soon show that P e A (I) if and only if for some minimal prime q,

P/q e A*(l+q/q).

Remark: We point out that A (I) is well behaved with respect to localization.

That is, P e A (t) i f and o n l y i f PS e A ( I s ) , S m u l t i p l i c a t i v e l y c l o s e d , S N p =~.

We have yet to show that A (I) ~ A (I). We choose to follow a path laid out

by Dan Katz [Kzl] which touches many useful ideas.

PROPOSITION 3.10. Let ~ be the Rees ring of R with respect to I. The follow-

ing statements are equivalent.

(i) P e A (I)

(ii) There is a prime p e A*(t-l~) with pNR=P.

 18

Proof: Since In = t - n ~ N R , any prime in ~(I) lifts to a prime in A*(t'l~).

Conversely, suppose p satisfies (ii), and write p = (t-n~:ct TM) with c e im .

Since the degree m component of t-n!R is (In+raN Im)t TM, and since pAR=P, we

have p = (In+m N ITM : c) = (In+m : c) since c e ITM. Clearly c ~ I n+m since P#R.

Thus n+m > O, and we have P e A(l,n+m) c_ A (I).

LEMMA 3.11. Let I be an ideal in a Noetherian ring R. Then ~ I n, n = 1,2,3, ...

is the intersection of those minimal primes q such that l+q@R.

Proof: First assume that R is a domain. Let V be any D.V.R. containing R.

Then I n ~ InV c InV = Inv since InV is principal and V is integrally closed.

Since NlnV= 0, NI n= 0 as desired.

In general, say ql'''" qr' q r + l ' ' ' " qs are the min~nal primes of R with

l+qi#R exactly when i = 1,2,...,r. For these i we have [ ( N I n) + q i ] / q i c_

N [(I+qi)/qi ]n (noting that I + q i is proper) which by the domain case shows that

n I n c_ ql N...N qr " Now let x e ql N...N qr " For any n > 0 and for any

j =r+l,...,s, we have In + q j =R. Thus there is a y e qr+l D...N qs and an

z e In with z + y = i. Therefore x = zx+yx. Now yx is in the radical of R,

h e n c e in •I n . As zx e I n c I n , we have x e In for all n > O.

LEMMA 3.12. Let I, P and q be ideals in a Noetherian ring R with q a

minimal prime and with P a prime minimal over I + q° Then there is an n ~ I

m m
such that for any m>n and any ideal J with I cJ~-I , we have PeAss(R/J).

Proof: Localizing at P, we have pk ~ l + q for some k > O. As q is minimal,

there is an x @ q with xq n = 0 for large n. Since R is local, the previous

lemma shows that we may choose n large enough to assure that x @ In , as well.

Let m >__ n and let ITM c:::J c ITM. Now p2mk c_ (l+q) 2m c_ I m + q TM, so that

p2mkx c_ I m x + q m x = I m x c_ Im c J. Since x @ Im , x ~ J. Thus p2mk consists of

zero divisors modulo J. As P is maximal, P £ Ass (R/J).

COROLLARY 3 •13. Let I, P and q be as in the previous lermna. Then P e A (I)

and P ¢ A (I).
 17

Proof: Obvious.

Comparing this corollary to Proposition 1.14 the question arises whether we

can weaken the above hypothesis to just q is a prime divisor of zero. The

answer is no. In [FR] there is an example of a complete 2-dimensional local

~,M) such that every minimal prime has depth 2 and there also exists a prime

divisor of zero, q, with depth q = I. Choose a e M -q. Then M is minimal

over a R + q. However M ~ ~*(aR) since if it were, then by Lemma 3.7 ( and Prop-

osition 3.4) there would be a minimal prime p with M/p ~ ( a R + p / p ) . Since R

is complete, R/p is quasi-unmixed and satisfies the altitude formula (see the

Appendix). Our next lermma thus shows that height M/p = I, contradicting that

depth p = 2.

LEMMA 3.14. Let a ~ 0 in a Noetherian domain R which satisfies the altitude

-@
formula. Then A (aR) = [ P primela e P and height P = I } .

Proof: One inclusion is obvious. Suppose that P e A (aR). By Lemma 3.2,

contains a height 1 prime p lying over Po Since R satisfies the altitude

formula, height P = height p = i.

Lemma 3.14 will be greatly strengthened in Chapter 4.

LEMMA 3.15. Let T be a faithfully flat ring extension of R. If I is an ideal

of R then I-~N R = ~.

Proof. Let ~ = R [ t -l,It] and ~' = T i t "l,It]. We have ~= t-l~NR and

I-~= t'l~ ' N T, so that it will suffice to show that t-l~ ' N ~ = t-l~. Therefore,

since ~' is a faithfully flat extension of ~, it will be enough to prove the

statement of the lemma in the special case that I = bR is principal. For this,

suppose that x e bTNR. We easily see that for some n,

x n e (bxn-iT+b2xn-2T+..,+bnT) NR= (bxn-l,...,bn)TNR= (bxn-l,o..,bn)R, since T

is a faithfully flat extension of R. This clearly shows that x e bR.

 18

PROPOSITION 3.16. Let (R,M) be a local ring with completion R , Let I be an

-. . -@ .
ideal of R. Then P e A (I) iff and only if there is a P e A ( I R ) with
.
P N R = P,

Proof: By Lemmaa 3.15, we see that any prime in A (I) lifts to a prime in

A (IR). Thus suppose that P e A (IR) and let P=P NR. Also let
* * -i *
~ = R [ t - l , lt] and ~ =R [t ,IR t]. By Proposition 3.10, there is a

p* e ~ ( t - l f ~) with p* N R . = P . . By Lermna 3.7, for some minimal prime q * ~_ p .

we have p*/q~ e A (t ~ +q /q ). Now R being complete implies that R*/q*NR*

satisfies the altitude formula, as does its finitely generate extension ~ /q . By

p*/q* *
Lemma 3.14, we see that has height I. Thus p is minimal over
)6~* -m *
t'l.cR*+q * Since for any m > O, t-m~ * ~ (t-m~ c t ~ , we see by Lemma 3.12
* )~* *
that for large m, p is a prime divisor of (t-m~ Now R c R is a flat ex-

tension. Thus so is ~ ~ ~ . Therefore p N~ is a prime divisor of t'mc~. By

Proposition 3.10, P = (p N ~ ) N R is in A (I).

PROPOSITION 3.17. Let I be an ideal in a Noetherian ring R. Then A (1)~A*(1).

In fact, if P e A (I) then either P is minimal in R or P e B (I).

Proof: Assume that P e A (I) is not minimal, and localize at P. By Proposition

. --. . * *
3.16, we see that P e A (IR). Take p and q as in the proof of that propo-

sition. Since p• is minimal over t -i ~ * + q * , by Corollary 3.13, p* e A*(t-l~ *)

As ~ c ~ is a flat extension, p • N ~ e A* (t-l~). Now t -I is regular in !R,

and so we easily see that p N is a prime divisor of t-l~. Also, in the proof

of P r o p o s i t i o n 3.16 we s a w t h a t p N~ was a prime divisor of t-m~ for large m.

*
Since P = (p n 2 ) N R is not minimal, by Lenmna 3.8, It ~ p * N ~ . Proposition 1.15

tells us that P E B (I). Finally, since any minimal prime of R containing I

is in A (I), we see that A (I) C A (I).

PROPOSITION 3.18. Let I be an ideal in a Noetherian ring, and let ~ =R[t-l,lt].

The following statements are equivalent.

i) P e A (I).

ii) There is a p e ~(t-l~) with pNR=P.

 19

iii) There is a prime divisor p' of t'l~ with p' A R = P . (~ is the

integral closure of N in its total quotient ring.)

iv) There is a minimal prime q of R with q ~ P and P/q e A*(l+q/q).

Also, if R is local with completion R , the above are equivalent to

, --, , *
v) There is a P e A (IR) with F NR=P.

Proof: We already have (i) <==> (ii) <==> (v). By Lermma 3.7 and Proposition 3.4, we

have (i) ~ (iv). We now show (iv) ~ (i), proceeding in three steps. First, assume

that I is principal and that R/q satisfies the altitude formula. Then I + q/q

is principal, and by Lemma 3.14 we have that height P/q = I (or else I + q/q = 0

in which case P =q is minimal, making (i) trivially true). Thus P is minimal

over l+q, and so P e A (I) by Corollary 3.13. For step two, we only assume

that I is principal. Now we may localize at P, and assume that R is local.

By (i) ~ (v) and the fact that (R/q)* ~ R /qR , we have that
* * -, , * *
P /qR e A (IR + q R /qR ). Now by (i) ~ (iv) and the fact that minimal primes of

R* /qR* have the form q*/qR* with q* a minimal prime of R*, we see that there

is such a q* with p*/q* e A--*(IR* + q * /q * ). Since I is principal and R*/q*

satisfies the altitude formula, by the first step we have that P e A (IR). By

(v) ~ (i) we have P e A (I).

For the final step, we consider the general case. Let N0 be the Rees ring

of R/q with respect to I+q/q. Let q+=qR[t-l,t] AN, which is a minimal prime

of N, and notice that N/q+ ~ ~0 " Since P/q e A (I+q/q), (i) ~ (ii) shows

that there is a prime p of N with p/q+ e ~ ( t - i N + q + / q + ) , and with

p/q+NR/q=P/q, so that pNR=P. Step two of our argument shows that

p e A*(t-IN), and so by (ii) ~ (i) we have P e ~(I). This proves that (iv)~(i).

For (ii) ~ (iii) we observe that t-nN= t-n~N R so that any p e ~*(t-IN) can

be lifted to a prime divisor p' of t'n~ for some n. As t -I is regular in

~, p' is also a prime divisor of t-l~. Finally, for (iii) ~ (ii), let p' be a

prime divisor of t-l~. Then height p' = I. Let q' be a minimal prime of

contained in p'. Then q = q' N ~ is a minimal prime of N contained in p = p' NN.

 20

Since height p'/q' =I, Proposition 1.7 shows that p/q e ~*(t'l~+q/q). By
-* -I
(iv) ~ (i), p e A (t ~), so that (iii) ~ (ii).

By now, our next result, due to Ratliff, is straightforward. However, it

motivated many of the ideas in the subject (such as Proposition 1.7) and so warrants

mention.

PROPOSITION 3.19. JR6] Let (R,M) be a local domain with completion R and in-

tegral closure [. The following are equivalent

(i) R contains a depth I minimal prime.

(ii) M e A (I) for every ideal I # 0.

(iii) M e Ass(R/(a)) for some a#O.

(iv) R contains a height i maximal prime.

(v) There is an n > 0 such that for every ideal 0 # I c_M n, M e Ass(R/l).

(vi) There is an n > 0 such that for every ideal 0 # I c M n, M e Ass(R/T).

Proof: (i) ~ (ii). Let I4 0 and let M be the maximal prime of R . Also

let q be a depth 1 minimal prime of R . Then M /q is minimal over

IR * + q * /q * , so that M * /q * e A * ( I R * + q*/q*). By Proposition 3.18 (iv) ~ (i) and

(v) ~ (i), we have M e A (I).

(ii) ~ (iii) is obvious.

(iii) ~> (i) by Proposition 3.4, we have M e A (aR). By Proposition 3.18 used

twice, there is a minimal prime q* of R* with M*/q* e X* (aR* +q*lq*)l. Now

R*/ q * satisfies the altitude formula, so Lemma 3.14 shows that depth q * = i.

(iii) ~ (iv) is by Lerm~a 3.2.

(iv) ~ (ii) is by Proposition 3.5.

(iv) ~ (v) is by Proposition 1.7.

(v) ~ (vi) is immediate since I c Mn implies T c M n.

(vi) ~ (iii) is obvious.

 21

The following proposition and its applications are due to K. Whittington [W ].

PROPOSITION 3.20. Let I= (a I ,..., an) be a nonzero ideal in a Noetherian domain

R. Let I c p with P prime. The following are equivalent.

(i) P e A (!).

(ii) For some ai, there is a prime Q of A =R[al/ai,... , an/a i] with

Q e A (aiA) and QNR=P.

(iii) For some ai, the integral closure of R[al/ai,... , an/ai] contains a

height 1 prime lying over P.

Proof: (ii) <~:~(iii) easily follows from Proposition 3.19 after the appropriate

localization.

(i) ~ (ii). Let ~=R[t-l, lt]. By Proposition 3.10, there is a p e ~(t-l~{) with

pNR=P. By Proposition 3.19, there is a height I prime p' of ~ with

p'N~R=p. Let (V,N) = (c~p,, pp,), which is a D.V.R. By eermna 3.3, It ~_ p, Thus

pick a.l such that a.tl ~ p" With A = R [ a l / a i,..., an/ai] , we have A c_ V, since

aj/ai=ajt/ai t and a.tl e V - N . Let @ = A I t - l , a i t] be the Rees ring of A with

respect to a.A.l Clearly A c @ c ~ c V. Also ~ c_ @, since a.tj= (aj/ai)ait.

Thus fRc @ c~2V. Since N = pp, , obviously NN ~ is a height i prime of ~ con-

taining t -I Proposition 3.19 shows that NN @ e ~(t-l~). By Proposition 3.10,

NNA e A (aiA)° Letting Q=NNA, we see that (i) ~ (ii).

(ii) ~ (i). Let a i, A and Q be as in (ii). By Proposition 3.10, (~=A[t'l,ai t]

contains a prime q e ~(t-l~) with qNA=Q. By Lemma 3.3, ait ~ q. Now Propo-

sition 3.19 shows that ~ contains a height I prime q' lying over qo Let

(W,K) be the D.V.R. -- , q'q,). Now

(@q, ~ = [t-l, It] c_C
- @, so that W is a locali-

zation of an integral extension of a finitely generated extension of ~. Clearly

the transcendence degree of W/K over @/q is 0. Now (~ is obtained by adjoin-

ing aj/ai= ajt/ait to ~, and a.tl ~ q" Thus the transcendence degree of @/q

over ~/qN~ is 0. Therefore, the transcendence degree of W/K over ~/qN

 22

is 0. By Lemma 3.1, height K A Y = i. Since t -I e K Q ~, Proposition 3.19 shows

that qN[R=KN~ e ~(t-l~). Finally, p = ( q N ~ ) NR, and Proposition 3.10 shows

that P e A (I).

PROPOSITION 3.21. Let I~ 0 be an ideal in a Noetherian ring R, and let X be

an indeterminate. Then A*(IR[X])= [PR[X]IP e A*(1)}.

Proof: First assume that R is a domain, and let P be a prime containing I.

Let I = (aI ,..., an, b), and let A=R[al/b,...,an/b]. Clearly A contains a

height i prime lying over P if and only if A[X] contains a height I prime

lying over PR[X]. Thus Proposition 3.20 easily shows that P e A (I) if and only

if PK[X] e A (IR[X]).

Now let Q be prime in R[X], with QNR=P, but Q~PR[X]. To complete

the domain case, we must show that Q ~ ~(IR[X])° Suppose, contrarily, that

Q e A (IR[X]). We may assume that R is local at P, so that Q = (P,f(X)) with

f(X) a monic polynomial. By Proposition 3.20, we may write I= (aI ,..., an, b)

such that if A=R[al/b,...,an/b] then A[X] =R[x,al/b,...,an/b] contains a

height ! prime q lying over Q. Let p=qOA. Since qNR =QNR=P, pNR=P

so that p # 0. Now since height q = i, we must have q = pA[X]. However

f(X) e Q e q and f(X) is monic so that f(X) ~ pA[X]. This contradiction com-

pletes the domain case.

For the general case, we have Qe~(IR[X]) if and only if Q / q e ~ ( I R [ X ] + q / q )

for some minimal prime q of R[X], by Proposition 3.18. Now q has form

q=pR[X] with p a minimal prime in R. Since the isomorphism R[x]/q m R/p[X]

carries IR[X]+q/q to (l+p/p)R/p[X], and since by the domain case just discussed,

the primes in A ((l+p/p)R/p[X]) have the form (P/p)R/p[X] for P/p e A (l+p/p),

we see that Q e A (IR[X]) if and only if Q/q is isomorphic to (P/p)R/p[X] for

some P/p e ~ ( l + p / p ) . Finally P/p e A (l+p/p) if and only if P e A (I) by

P r o p o s i t i o n 3.18. Because Q/q is isomorphic to (P/p)R/p[X] implies Q=PR[X],

we are done.
 23

PROPOSITION 3.22. Let R c T be an integral extension of Noetherian rings. Let

I be an ideal in R° If P e A (I) then there is a Q e A (IT) with QNR=P.

If also every minimal prime of T contracts to a minimal prime of R, then

Q e A (IT) implies QNR e A (I).

Proof: First suppose that R and T are domains, and let P be prime in R.

We must show that P c A (I) if and only if there is a Q e A (IT) with QNR=P.

If P s A (I), by Proposition 3.20 we may write I = (a I ,..., a n, b) such that

R[Ib -I] contains a height i prime p lying over P. Now R[Ib -I] ~ T[ITb -I]

is an integral extension. Thus there is a height I prime q of T[ITb -I] lying

over p. Let Q=qN T. By Proposition 3.20, Q e A (IT). Also QNR=P. The

converse of the domain case is similar, except it needs the additional fact that

R[Ib -I]- ~ T[ITb

-I]- satisfies going down, so that height 1 primes contract to

height i primes.

In the general case, suppose that P c A (I). By Proposition 3.18 (i) ~ (iv),

P/p e A (l+p/p) for some minimal prime p of R. Lift p to a prime q of T,

which must be minimal. Applying the domain case to R/p ~ T/q, we find a

Q/q e ~ ( I T + q / q ) with Q/qNR/p=P/p. Thus QNR=P and by Proposition 3.18,

Q e A (IT). If every minimal prime of T contracts to a minimal prime of R, the

reverse works similarly.

Remarks: (a) The condition on minimal primes above is required. It is easy to

find R ~ T as above with R a domain but q a minimal prime in T with

qNR#0. Now q e A (OT) but qNR ~ A (OR).

(b) The above result fails for A (I). To see this, let (R,M) be a 2-dimensional

local domain which satisfies the altitude formula, but which is not Cohen-Macaulay,

and let R be Noetherian. (Such a situation is easily constructed.) For O# aeM,

we have M e A (aR) since R is not Cohen-Macaulay. Suppose N is a prime of

lying over M. In view of the altitude formula, we have height N = height M = 2.

As R is a Krull domain, N cannot be a prime divisor of anR for any n. Thus

N @ A (aR).
 24

Let R be a N o e t h e r i a n domain and I~ 0 an ideal. If P e A (I), then

Proposition 3.20 tells us t h a t f o r some 0~b e I, R [ I b - l ] contains a height 1

prime Q lying over P. Let (V,N) be t h e D.V.R. (R[Ib-1]Q, QQ).

LEMMA 3.23. Let R , I, P , b , Q and V be as above. If J#'0 is an ideal of R

with JV c IV and (J : I ) V = (JV : IV), then P e A (J).

Proof: Since JV c IV, J c I V N R c:~ N f ~ R = p . Also JV c IV are both principal

ideals of V and so J V = (JV : IV)IV. Note that IV = b V . If we choose c e J and

d e (J : I) such that JV = cV and (J : I ) V = dr, then c V = J V = (JV : IV)IV =

(J : I ) V I V = d b V . Since J ~ JV=cV=dbV, J(db) -I ~ V. Also, since b e ! and

d e (J : I), bd e J. Furthermore, Id ~ l(J : I) so that Ib -I ~ l(J : l)(bd) -I ~_

J(bd) -I. Thus R[Ib -I] ~ R[J(bd) -I] ~ V, so that R[Ib -I] ~ R[J(bd) -I] ~ V.

Since N N R [ I b -I] = Q has height I, clearly N n R [ J ( b d ) -I] has height I. Since

this prime lies over P, and because we already have J ~ P and bd e J, Propo-

sition 3.20 shows that P e A (J).

If P e A (I) and if I c J c T, then obviously P e A (J). The next coro-

llary goes a little further, since T ~ IV N R .

COROLLARY 3.24. Let R , I ,P and V be as in Lermna 3.23. If J is an ideal of

R with I c J c IVnR, then P e A (J).

Proof: Obviously IV=JV and (J : I)V = RV = V = (JV : IV).

COROLLARY 3.25. Let P e A (I). Then there is an n > 0 such that if J is any

ideal with I c J c I+pn, then P e ~(J).

Proof: First assume that R is a domain. By Proposition 3.20, we can find b, Q

and V as in Lermna 3.23. Since V is a D.V.R., for some n we have pnv c_ IV.

Now pn c pnv c IV. Thus for J as above, I c J c IVf]R, and so P e A*(J) by

Corollary 3.24. The general case follows by P r o p o s i t i o n 3.18 ((i)~>(iv)) and the

fact that pn+q/q ~-- (p/q)n for q ~ P with q a minimal prime.

 25

PROPOSITION 3.26. Let I ~ P be ideals of R with P prime. Assume dim R > 0.

The following are equivalent.

(i) P e A (1).

(ii) P e A (IJ) for any ideal J with height J > O.

(iii) P e A (Ic) for any element c not contained in any minimal prime.

(iv) There exists an element c not in any minimal prime, with P e A (Ic).

Proof: (ii) ~ (iii) is immediate, as is (iii) => (iv) since dim R > 0. W e will

prove (i) ~ (ii) and (iv) =~ (i) for R a domain. The general case is then straight-

forward using P r o p o s i t i o n 3.18 ((i) ~ ( i v ) ) .

Assume R is a domain, and let P e A (I). Let J be any nonzero ideal.

Since P e A (I), Proposition 3 . 2 0 shows t h e e x i s t e n c e of a b, Q and V as i n

L e m m a 3.23. As IV is principal, (IJV : IV) = (JV : V) = JV, so that (IJ : I)V ~_

(IJV : IV) = JV ~ (IJ : I)V. Thus (IJ : I ) V = (IJV : IV). As IJV ~ IV, Lemma 3.23

shows that P e A (IJ). Thus (i) :::::>(ii). For (iv) ~ (i), suppose c~0 and

P e A (Ic). Let I = (a I , ..., an). By Proposition 3.20, for some i,

R[alc/aic,...,anC/aic] contains a height 1 prime lying over P. As t h i s ring

equals R[al/a i,..., an/ai] , that proposition shows P e A (I).

In C h a p t e r 4, we w i l l g i v e an e x a m p l e w h e r e P e A (IJ), but P ~ A (I) and

P~A (J).
 CHAPTER IV: A Characterization of A ( 1 )

In this chapter we investigate a fundamental characterization of A (I) for

ideals in a Noetherian ring which is locally quasi-unmixed (see Appendix). Recall

that if I is an ideal in a local ring (R,M), then ~(I), the analytic spread

of I, can be characterized in various ways. If f(n) is the minimal number of

generators of In , then there is a polynomial P(X) such that P(n) = f(n) for

all large n (the Hilbert polynomial), and ~ (I) = deg P +I. If ~ = R [ t -I, It] and

if ~ = ... + Rt - 2 + R t - l + M + I t + 1 2 t 2 +... , then ~(I) equals the height of

~/(t-l,M)~. Finally, if R/M is infinite, so that I has a minimal reduction,

then ~(I) is the size of a minimal basis of a minimal reduction of I. Recall also

that for a Noetherian domain R, being locally quasi-ur~nixed is equivalent to satis-

fying the altitude formula.

PROPOSITION 4.1. [M4] Let I be an ideal in a Noetherian ring R and let P be

a prime containing I. If height P = ~ ( ~ ) , then P e A (I). If ~ is quasi-

unmixed, then the converse is also true.

Proof: We first treat the case that R is a domain. We localize at P. Since

~(I) = ~(IR(X)) (note: R(X) =R[X]pR[X]), by Proposition 3.21 we may assume that

R/P is infinite. Let J be a minimal reduction of I. Since im= jm for all m~

obviously A (1)=A (J), and we may take l=J= (al,... , an) with ~(1)=n. Now

a I ,..., an are analytically independent. Letting A=R[al/an,..., an_l/an] we

have that PA is a prime in A with height PA <_ height P - n + i. Suppose now

that ~(I) = n = h e i g h t P. Then height P A = I so that A has a height 1 prime

lying over P. Therefore, by Proposition 3.20 we now see that P e A (I). Con-

versely suppose that P e A (I) and that (R,P) is quasi-unmixed, so that it sat-

isfies the altitude formula. By Proposition 3.20 we have that A contains a

height I prime q lying over P. As A is finitely generated over R, A also

satisfies the altitude formula, and so height(qN A ) = height q = I. Since PA is

clearly contained in q N A, we have PA = q N A. Thus height PA = I. However, the

altitude formula also shows that height PA = height P - n + I. Thus height P = n = ~ (I).
 27

This completes the domain case. Before doing the general case, we require two

lermnas.

LEMMA 4.2. Let I be an ideal in a local ring OR,M). For any minimal prime q,

I(l+q/q) <_ ~(I), and equality holds for some such q.

Proof: Since the minimal number of generators of In+q/q does not exceed the min-

imal number of generators of In , the characterization of analytic spread in terms

of Hilbert polynomials gives the inequality. Now in ~R, the Rees ring of R with

respect to I, let N = ...+ R t ' l + M + I t + 1 2 t 2 + . . . and J = ...+ R t - I + M + M I t +

Ml2t 2 +... so that ~(I) =height(N/J). If this number is n, let J c P0 c

PI c . . . c Pn = N with Pi prime. Let Q be a minimal prime of ~ with Q ~ PO "

If q=Q N R it is straightforward to verify that q is a minimal prime of R and

that Q = ...+ q t - l + q + ( I N q ) t + ( 1 2 N q ) t 2 + .... Clearly N/Q modulo J+Q/Q has

height n. Now if ~' is the Rees ring of R/q with respect to l+q/q, and if

N' = ...+ R/q t ' l + M / q + (l+q/q)t+ (12+q/q)t 2 +... and J' = ...+ R/q t - l + M / q +

(Ml+q/q)t+ (Ml2+q/q)t 2 +... then ~(l+q/q) =height N'/J'. However ~/Q ~ ~', the

isomorphism taking N/Q to N' and J+Q/Q to J'. Thus height N'/J' = n = ~ ( 1 ) .

LEMMA 4.3. (Rees) Let I be an ideal in a local ring 0R,M). Then ~(1)<heightM.

Proof: By the previous lerm~a, for some minimal prime q, ~(I)=~(l+q/q). Since

height M/q <_ height M, we may assume R is a domain. We may also assume that

R/M is infinite, and let (aI , ..., an) be a minimal reduction of I, with

n = ~(I). Thus a I , • •., an are analytically independent. Applying the altitude

inequality to the primes M and MR[al/a n,,.., an_l/an] , since the transcendence

degree of R]al/a n,..., an_l/a n ] modulo MR[al/a n .... , an.l/an] over R/M is n - I,

we see that height M >__height MR[al/an,..., an_l/an] + (n-l) >__n = ~ ( 1 ) .

Proof of Proposition 4.1: (continued) In general, suppose that height P equals

~(~). We may assume that R is local at P~ By Lermma 4.2, pick a minimal prime

q of R with ~(1)=~(l+q/q). By Lermna 4.3, we see that ~(l+q/q)=height P/q.

By the domain case, P/q e A (l+q/q), so that by Proposition 3.18, P e A*(1).

 28

Finally suppose that P e A (I) and that (R,P) is quasi-unmixed. By Propo-

sition 3.18, for some minimal prime, P/q e A (l+q/q). By the domain case,

height P/q = ~ (l+q/q). As R is quasi-unmixed, height P/q = height P. By Lemmas

4.2 and 4.3, height P = ~ ( 1 ) .

COROLLARY 4.4. Let (R,M) be a local ring with completion R . If I is an ideal

of R, then M e A (I) if and only if for some minimal prime q of R ,

£(IR +q /q') =height M /q .

Proof: By (i) => (v) and (i) ~ (iv) of-Proposition 3.18, M e A (I) if and only if

for some minimal q of R , M /q e A (IR +q /q ). As R /q is quasi-unmixed,

we invoke Proposition 4.1 to complete the proof,

Let R be locally quasi-unmixed. Then Proposition 4.1 shows that P e A (I)

if and only if height P = ~ ( ~ ) . This property characterizes locally quasi-unmixed

rings. That is, if for all I in R, P e A (I) if and only if height P = ~ ( ~ ) ,

then R is locally quasi-unmixed. (In fact we can restrict ourselves to ideals in

the principal class, i.e. for which the height of I equals the minimal number of

generators of I.) We could prove this now, using Corollary 4.4. Instead, we defer

it until Chapter 5, when we will have the machinery to give a very brief proof.

EXAMPLE. We give the previously promised example of P e A (IJ) with P @ A (I)

and P @ ~(J). Let (R,P) be a 3-dimensional quasi-unmixed local domain. Let

a,b,c be a system of parameters. Let I= (a,c) and J = (b,c). Since ~(I) does

not exceed the minimal number of generators of I, ~(I) ! 2 < height P° By Propo-

sition 4.1, P @ ~*(I). Similarly P @ ~(J). Now IJ = (ab, ac,bc, c2), and since

ab/c 2 = (a/c)(b/c), we have R[IJ(c2) -I] =R[a/c,b/c]. Since a,b,c is a system

of parameters, PR[a/c,b/c] is a height I prime ideal. Thus R[IJ(e2) -I] con-

tains a height I prime lying over P, and so P e A (IJ) by Proposition 3.20

EXAMPLE. Let R equal K[X,Y,Z,W]/(XY-ZW) localized at (X,Y,Z,W)/(XY-ZW). With

primes denoting images in R, let P = (X',Z'). We will show that A (P) = {P}.

Let p e A*(P). Since R satisfies the altitude formula, height p = ~ ( P ) by

P
 29

Proposition 4.1. Since ~(Pp) does not exceed the minimal number of generators

of P p , height p <-- 2. Thus p ~ (X',Y',Z~,W'). Since P ~ p we must have either

Y' ~ p or W ~ @ p. Since X'Y' =Z'W', we therefore easily see that P is gen-

P
erated by either X' or Z' Thus height p = ~(Pp) = i, so that p=P°

Remark: The above example is quite interesting. P is not principal since if it

were, then (X',Y',Z',W') could be generated by three elements. Since

XY - Z W c (X,Y,Z,W) 2, this would give (X,Y,Z,W) generated by three elements.

Thus P is not principal. As R is normal [$2, Theorem i], ~(P) > I (since if

(P) = I with R normal, there is an a e P with aR ~ P ~ a R = a R ) . Thus ~(P)= 2

since P has two generators. However, the argument used above shows that ~(Pp) = i

for any height 2 prime p containing P.

LEMMA 4.5. Let (R,M) be a local domain with integral closure R, and let I# 0

be an ideal of R. Then ~(I) = i if and only if IR is principal.

Proof: By going to R(X) if necessary, we may assume that R/M is infinite.

Suppose that ~(I) = I. Then for some a e I, I = a R = a R N R . Therefore aR~ I~

TR ~ aR, showing that IR=aR. Conversely, suppose that iR=c~R for some a e R.

Choose c e R such that ca e R. Now cI=RN (Nel-V) as V ranges over all

valuation overrings of R. However clV = clRV= car. Thus c-~= R N ( N caV) =

R N [ca(NV)] = R N eaR= co~. This shows that ~(cl) = I° Thus ~(I) = i.

PROPOSITION 4.6. Let R be a 2-dimensional Noetherian domain with integral

closure R. Let 0 ~ I ~P be ideals with P prime. Then P e ~(I) if and only

if one of the following three conditions holds.

(i) P is minimal over I.

(ii) There is a height I prime in R lying over P~

(iii) ~(~) > i. (Equivalently, I~ S is not principal, with S = R - P, using

Lemma 4.5.)

Proof: If (i) holds, obviously P e A (I). If (ii) holds, use Proposition 3.19.
 30

If (iii) holds, since dim R = 2 we have ~ ( ~ ) = 2 = height P, by Lemma 4.3. Thus

Proposition 4.1 gives P e A (I).

-..-4¢
Conversely, assume P e A (I) and that (i) and (iii) both fail to hold.

Then c l e a r l y height P = 2 and £ ( I p ) = 1. Proposition 4 . 1 now shows t h a t does

not satisfy the altitude formula (i.e. is not quasi-unmixed). Since is 2-di-

mensional, the only way this can occur is for [ to contain a height i prime

lying over P.

For a 2-dimensional normal Noetherian domain, we find that A (I) always

equals A (I).

COROLLARY 4.7. Let R be a normal 2-dimensional Noetherian domain. Let I~ 0 be

an ideal. The following are equivalent for a prime P.

(i) P e A (I)

(ii) P e A (I)

(iii) Either P is minimal over I or I~ is not principal.

Proof: (ii) 4=>(iii) is immediate by Proposition 4.6. (ii) => (i) is by Proposition

3.17. Finally, suppose P e A (I), and assume that P is not minimal over I.

Then height P = 2. For large n, Pp is a prime divisor of q, which therefore

cannot be principal since R? is a Krull domain. Thus (i) ~ (iii).

Remarks: (a) The class of domains in which A (I) always equals A (I) will be

studied in a later chapter.

(b) The equivalence of (i) and (iii) in Corollary 4.7 was first proved by P. Eakin

[ME, Proposition 21]. We use Eakin's arguments in our next result.

PROPOSITION 4.8. Let (R,M) be a 2-dimensional local Cohen-Macaulay domain with

multiplicity e. If I is an ideal and M e A (I), then M e Ass(R/I n ) for all

n>e.

Proof: Let n > e and suppose that M ~ Ass(R/In). Then height I = I and R/I n

is Cohen-Macaulay. Now a result of Rees generalized in [ SI] shows that In can be

 31

generated by e elements. By [ES, Corollary I] I is prestable, which one easily

sees implies ~(I) = I. As R, being Cohen-Macaulay, satisfies the altitude formula,

Proposition 4.1 shows that M ~ ~(I). This contradiction proves the result.

Remark: We do not know if the hypothesis can be weakened to M e A (I) or the con-

clusion strengthened to M e Ass(R/In), n > e.

 CHAPTER V: Asymptotic Sequences

In view o f t h e e x i s t e n c e of A (I), and t h e c l a s s i c a l c o n c e p t o f an R - s e q u e n c e ,

it is n a t u r a l to consider a sequence of elements x 1 ,..., x such that

n
(x I ,..., x n) ~ R and for each i, x i is not in any prime ideal in A ((x I,.,., Xi_l)).

However, it follows easily from IN1, E x e r c i s e 13, p. 1 0 3 ] , that such sequences are

exactly R-sequences. Instead, we go t o A (I), and make t h e f o l l o w i n g definition.

DEFINITION. The sequence of elements x I ,..., x n in R is an asymptotic sequence

if (x I ,..., Xn) # R and if for each 1 < i < n, x. is not in any prime contained
l

in A ((x I ,..., Xi_l) ).

Remarks: (a) When i = i, we have x I ~ U{P e A*(0)} = U [ p l p is a minimal prime

of R}.

(b) If x I , ..., x n is an asymptotic sequence in R, and if (Xl, ..., x n) ~ P

with P prime, then x I ,..., Xn is easily seen to be an asymptotic sequence in Rp .

This notion of an asymptotic sequence was independently conceived by Rees, Rat-

liff, and the author. Rees obtained the first significant result [Rs2], which will

be discussed in the next chapter. The main characterization of asymptotic sequences

(Proposition 5.4) to be given in this chapter, first appeared in Ratliff's work JR9].

However, K a t z independently followed roughly the same trail [Kzl]. W e have borrowed

extensively from both of these.

LEMMA 5.1. Let (R,M) be a local ring with completion R . Let x I ,..., x n be a

sequence of elements in R. The following are equivalent.

(i) xI ..., x is an asymptotic sequence in R.

' n

(ii) x I ,..., x n is an asymptotic sequence in R .

(iii) x I +q,...,x n+q is an asymptotic sequence in R/q for every minimal prime

q of R.

Proof: This is an easy consequence of Proposition 3.18.

 33

LE}~IA 5.2. Let xI ..., x be an asymptotic sequence in R and let P be a

' n

prime m i n i m a l over (x I ,..., Xn). If q is any minimal prime contained in P,

then height P / q = n .

Proof: W e may assume R is local at P. By L e m m a 5.1, we may assume that q= 0

and R is a domain. W e induct on n, the case n = I b e i n g trivial. N o w for P

m i n i m a l over (x I ,..., Xn) , w e o b v i o u s l y have height P ~ n. Shrink P to a prime

Q m i n i m a l over (x I ,..., Xn_l). By induction, height Q = n - I. Since

__.u
Q e A (x I ,..., Xn.l) we have Xn e P -Q. Therefore height P > height Q = n - i.

Thus height P = n.

O u r n e x t result shows that in a quasi-unmixed local ring, x I ,..., x n b e i n g an

asymptotic sequence is e q u i v a l e n t to I = (x I , ..., x n) b e i n g an ideal in the prin-

cipal class (i.e. height l = m i n i m a l number of generators of I). This does not work

globally. Let R have two m a x i m a l ideals w i t h height M I = i and height M 2 = 3.

Choose xI , x2 , x3 such that h e i g h t ( x I , x 2 , x 3) = 3 but w i t h x I , x 2 e M I • Since

MI e A ((Xl)), x I , x 2 , x 3 is not an asymptotic sequence.

L E M M A 5.3. Let (R,M) be a q u a s i - u n m i x e d local ring. Then X 1 ~'''~ X n is an

asymptotic sequence if and only if h e i g h t ( x I ,..., x n) = n .

Proof: One direction is immediate from Lermna 5.2. Thus suppose that

h e i g h t ( x I ,..., Xn) = n . W e induct on n, the case n = i b e i n g trivial. Let Q

b e a p r i m e minimal over (x I ,..., Xn_l). Therefore height Q ~ n - I, so that

x ~ Q. As R is local there is a prime P m i n i m a l over (Q,Xn). W e have

n
height(P/Q) = i, and in v i e w of quasi-unmixedness, height P = h e i g h t Q + I ! n. Since

(x I ,..., Xn) ~ P, the h y p o t h e s i s gives height P h n. Thus height P=n and so

we have height Q = n - l . This shows that h e i g h t ( x I ,..., Xn_l) = n - l . B y induction,

x I ,.°., Xn_ I is an asymptotic sequence. N o w let p e A (x I ,..., Xn_l). By

P r o p o s i t i o n 4.1, height p = ~ ( ( x I ,o,., Xn_l)Rp) , and of course this analytic spread

does not exceed n - i. Therefore height p ! n - i . By hypothesis, clearly x n ~ p,

for any p e A (x I ,..., Xn_l). This shows that x I ,..., x n is an asymptotic

sequence.
 34

PROPOSITION 5.4. Let (R,M) be a local ring with completion R . Let

(x I , . • ., Xn) ~ R . Then x I , .. ., x n is an asymptotic sequence in R if and only if

J- *
height(x I ,..., X n ) R " + q / q = n for every minimal prime q of R .

Proof: Since R /q is quasi-unmixed, this is irmnediate from lemmas 5.1 and 5.3.

COROLLARY 5.5. Let Xl ''''' nX be an a s ~ p t o t i c sequence contained in the Jacobson

radical of R. Then any permutation of x I , • .., x n is also an asymptotic sequence.

Proof: It is easily seen that if xI ..., x is in the Jacobson radical, then it

' n

is an asymptotic sequence in R if and only if it is an asymptotic sequence in RM

for each maximal ideal M. The result is now easy using P r o p o s i t i o n 5.4.

In the following material we will be repeatedly discussing, for a prime P,

the m i n i m u m of the depths of minimal primes in (R~) • W e therefore introduce

notation.

DEFINITION. If P is prime in a N o e t h e r i a n ring R, let z(P) = min[depth q lq

is a minimal prime in (R~)*}.

PROPOSITION 5.6. Let I be an ideal in a Noetherian ring R. Let x I,..., x n

be an asymptotic sequence maximal among asymptotic sequences contained in the

ideal I. (That is, there is no Xn+ I e I with x I ,..., X n + I an asymptotic

sequence.) Then n=min[z(P)l P is a prime containing I} = m i n [ z ( e ) I P e A*(1)}.

Proof: Let P be a prime containing I. Now x I ,..., Xn is an asymptotic

sequence in ~ . By Proposition 5.4, for any minimal prime q of Rp,

height(x I , . . . , Xn) ~ + q / q = n . Thus depth q >_ n, so that z(P) >_ n. This gives

half of the first equality. Now since x I , ..., x n is a maximal asymptotic sequence

from I, there is a Q e A-* (x I ,..., Xn) with I c_ Q. Clearly x I ,..., x n is a

maximal asymptotic sequence from QQ in RQ , and by two uses of Proposition 3.18,

for some minimal prime q of ~ , /q e A ((x I ,..., Xn)R Q + q / q ) . By

 35

Proposition 4.1, height Q Q / q = ~ ( ( x I ,..., X n ) R Q + q / q ) ~ n. Thus z(Q) ~ n, proving

the first equality.

In order to prove the second equality, it will suffice to show that the Q

discussed above is in A (I), since we clearly have z(Q) = n. Therefore we com-

plete the proof of the present proposition by proving the following lemma.

LEMMA 5.7. Let x I,..., x n be an asymptotic sequence and let Q e A (x I,..., x n).

If I is an ideal with (xI,.,o, Xn) ~ I ~ Q, then Q e A (I).

Proof: We may assume that R is local at Q. Obviously x I ,..., x n is a maximal

asymptotic sequence in Q, and so by t h e f i r s t equality in Proposition 5.6 (applied

to the ideal Q) z(Q) = n. Thus R contains a minimal prime p of depth n. By

Proposition 5.4, height(x 1 ,..., Xn)R + p / p = n , so t h a t Q is minimal over

(x I ,..., Xn)R +p. Thus Q is minimal over IR +p. By Corollary 3.13,

Q e A (IR), and by Proposition 3.18, Q c A (I).

Remark: The analog of Lermna 5.7 fails for classical R-sequences. That is, if

x I,•,., x n is an R-sequence and if Q e Ass(R/(x I ,..., Xn)) and if

(xI,°.., Xn) ~_ I c Q, we cannot conclude that Q e Ass(R/I)•

Proposition 5.6 shows that all asymptotic sequences maximal with respect to

coming from I, have the same length.

DEFINITION. Let I be an ideal in a Noetherian ring R. The asymptotic grade of

I, gr I, is the common length of all maximal asymptotic sequences from I.

We remark that if S is a multiplicatively closed set with I~ S = ~, then

gr*(1) ! g r * ( ~ ) . This is straightforward.

We give several corollaries to Proposition 5.6. The first shows that Cohen-

Macaulay rings are to R-sequences as locally quasi-unmixed rings are to asymptotic

sequences.
 38

COROLLARY 5.8. The following are equivalent in a N o e t h e r i a n ring R.

(i) gr (I) = height I for every ideal I of R.

(ii) gr (M) = height M for every maximal ideal M of R.

(iii) R is locally quasi-unmixed.

Proof: (i) ~ (ii) is trivial. For (ii) ~ (iii), by Proposition 5.6, we see that

gr (M) = z(M) for all maximal M. Thus height M = z(M) so that ~ is quasi-

unmixed by definition. Thus R is locally quasi-unmixed, and (iii) holds. Finally,

to show (iii) ~ (i), in a locally quasi-unmixed ring, height P = z(P) for all

primes P. Therefore by Proposition 5.6 we easily have (iii) ~ (i).

COROLLARY 5.9. Let ~,M) be a local ring. The following are equivalent.

(i) R is quasi-unmixed.

(ii) There is a system of parameters for R w h i c h is an asymptotic sequence.

(iii) Every system of parameters for R is an asymptotic sequence~

Proof: (i) ~ (iii) is irmmediate by Lemma 5.3, and (iii) ~ (ii) is trivial. For

(ii) ~ (i), let x 1 , ..., xn be a system of parameters for R which is also an

asymptotic sequence. Since x I ,..., Xn is a system of parameters, height M = n .

Since x 1 ,..., x is an as3nnptotic sequence, (which is obviously maximal with

n
respect to coming from M), gr M = n. Since height M = gr M = z(M), the last equality

by Proposition 5.6, we s e e that R is quasi-unmixed.

COROLLARY 5.10. Let x I, ..., Xn be an asymptotic sequence and let the prime P

be minimal over (x I ,..., Xn). Then ~ is quasi-unmixed.

Proof: Obviously x I ,..., x n is a maximal asymptotic sequence in Pp . Thus

z(P) = Z(Pp) = n. However, by Lermna 5.2, we also see that height P = n. Thus

height P = z(P), and we are done.

W e now give the promised converse of Proposition 4.1.

 37

COROLLARY 5.11. Let the Noetherian ring R have the property that for any ideal I

in the principal class and any prime P, P e A (I) implies height P = ~ ( ~ ) . Then

R is locally quasi-unmixed.

Proof: Assume that R is not locally quasi-unmixed. Then for some prime Q, RQ

contains a minimal prime of depth less than height Q. Thus, using Proposition 5.6,

we have m = gr Q <__ z(Q) < height Q. Let x I , ..., x m be a maximal asymptotic

sequence coming from Q. By Lenmaa 5.2, I = (x I , ..., Xm) is in the principal class.

Since x I ,..., Xm is a maximal asymptotic sequence coming from Q, there is a

prime P e A (xI , ..., Xm) with Q ~ P. Now height P >__ height Q > m ~ ~ ( ~ ) ,

since analytic spread does not exceed the minimal number of generators. We have

now proved the contrapositive of the result.

As another corollary, we get a strengthening of Proposition 3.19 (i) ~ (ii).

COROLLARy 5.12. Let P be a prime ideal. If I is an ideal contained in P and

if gr I = z(P), then P e A (I).

Proof: Notice that by Proposition 5.6, z(P) = gr (Pp). Call this number n. Let

x I ,..., Xn be a maximal asymptotic sequence from I. Clearly x I ,°.°, Xn is also

a maximal a s y m p t o t i c s e q u e n c e from Pp since g r Pp = Z(Pp) = n. Thus

Pp e A ((x I ,..., Xn)Rp) , so that P e A ((x I,..., Xn)R). By Lemma 5.7, P e A (I).

LEMMA 5.13. Let x I ,..., Xn be an R-sequence from R. Then x I ,..., x n is an

asymptotic sequence. In p a r t i c u l a r , gr I ~ gr I ~ h e i g h t I .

Proof: It follows easily from [KI, Exercise 13, p. 103] that if P cA (xI ,..., x i)

for some i, then P e Ass(R/(x 1 ,..°, x i ) ) ° Thus, u s i n g P r o p o s i t i o n 3.17, an R - s e q u e n c e

is easily seen to be an asymptotic sequence. Therefore gr(I) ~ gr (I) is obvious.

Now for any prime P, clearly z(P) ~ h e i g h t P, and so by P r o p o s i t i o n 5 . 6 ,

gr (I) ~ height I.

Our next result strengthens Proposition i.I0.

 38

PROPOSITION 5.14. Let T be an integral extension of R and let

q0 c ql c...c qn be a saturated chain of primes in T. Suppose that q0NR is a

minimal prime of R. Let P=qnNR" Then gr P ~ gr P ! n.

Proof: We already have the first inequality. Now asymptotic

* *
upon localization. Thus gr P ~ gr Pp = z(P). We will show that z(P) ! n. We may

assume that R is local at P. Letting P=q0NR' R/p ~ T/q0 is an integral ex-

tension, and T/q0 contains a maximal chain of length n. Thus, as discussed in

the Appendix, (R/p) contains a minimal prime of depth n. Now (R/p) = R /pR ,

and minimal primes in R~/pR ~ have form p*/pR* with p a minimal prime of R .

Thus R contains a minimal prime of depth n, so that z(P) ~ n.

PROPOSITION 5.15. Let pc_Q be prime ideals in a Noetherian ring. Then

z~) < z(e) +z(Q/P).

Proof: (~) * contains a minimal prime of depth z(P), and (RQ/PQ)* contains a

minimal prime of depth z(Q/P). Therefore, a result proved in the Appendix shows

that (RQ) contains a minimal prime of depth z(P) +z(Q/P). Thus z(Q) ~ z(P) +

z(Q/P) °

PROPOSITION 5.16. Let I be an ideal of R and let a be an element of the

Jacobson radical of R. Then gr I ~ gr (l,a) ! gr I + I .

Proof: The first inequality is obvious from the definition of asymptotic grade. We

now use Proposition 5.6 to find a prime P containing I, with gr I = z(P). If

(l,a) ~ P then gr I ~ gr (l,a) ~ z(P) = gr I, and we are done. If (l,a) ~ P

then a ~ P, but since a is in the Jacobson radical, there is a prime Q minimal

over (P,a). Thus height Q/P = I. Obviously z(Q/P) = I, and by Proposition 5.15

z(Q) ~ z(P) + I. Therefore Proposition 5.6 shows that gr (I, a) ~ z(Q) ~ z(P) + I =

gr I + i.

In Proposition 5,16, the condition that a be in the Jacobson radical is needed.

Let R be a domain satisfying the altitude formula and having two maximal ideals

M and N, with height M = I and height N = 3. Let I =MAN and a ~ N-M.

 39

Since z(M) = i and z(N) = 3, clearly gr (I) = i while gr (l,a) = 3. This example

also shows that in the next corollary, the local condition is needed.

COROLLARY 5.17. Let I be an ideal in a local ring (R,M), and suppose that

gr I < gr M. Then there is a prime P containing I with gr P = gr I + l .

Proof: Let x I ,..., x n be a maximal asymptotic sequence from I. Since gr M > n,

there is an Xn+ 1 e M with x I ,..., Xn+ 1 an asymptotic sequence. By Proposition

5.16, gr (I,Xn+ I) = n + l . Thus x I ,..., Xn+ 1 is a maximal asymptotic sequence

from (I,Xn+l) and so there is a P e A (x I ,..., Xn+ I) with (I, Xn+ I) ~ P.

P is as desired.

PROPOSITION 5.18. Let I = (al,..., an). Then gr (I) ~ n. If gr I = n , then I

can be generated by an asymptotic sequence (necessarily of length n).

Proof: By Len=na 5.13, gr I <_ height I, which clearly cannot exceed n. The proof

of the rest of the result is analogous to the proof of [KI, Theorem 125].

The example [KI, Example 7, p. 102] shows that in the above proposition, if

gr I = n, not every minimal basis of I is an asymptotic sequence. However, if I

is in the Jacobson radical, it is true.

PROPOSITION 5.19. Let I = (a I ,..., a n ) be in the Jacobson radical of R. Then

is an asymptotic sequence if and only if gr (I) = n.

a I , .-., a n

Proof: One direction is obvious. Suppose gr l = n . In order to show that

is an asymptotic sequence in R, w e easily see that it is enough to

aI . . . . . an
show t h e y form an a s y m p t o t i c s e q u e n c e i n RM f o r each maximal i d e a l M. Now

n = gr I ~ gr ~ ~ n by Proposition 5.18. Thus we may assume R local at M.

P r o p o s i t i o n 5.18 shows t h a t I = (b 1 , . . . , bn) with b1 ' ' ' ' ' nb an a s y m p t o t i c

sequence. Now by P r o p o s i t i o n 5.4, for any minimal prime q of R ,

height(bl,..., bn)R + q / q = n . Thus height(al,... , an)R + q / q = n andPropositionS.4

shows that a~ ,..., a is an asymptotic sequence.

£ n
 40

C O R O L L A R Y 5.20. Let I = (aI ,..., an) and J = (b I ,..., h n) be two ideals in the

J a c o b s o n radical of R° Suppose Rad I = R a d J. Then a I ,..., an is an asymptotic

sequence if and only if b I ,..., bn is an asymptotic sequence.

Proof: P r o p o s i t i o n 5.6 shows that gr I = g r J. The result is now irmnediate from

the p r e c e e d i n g proposition.

The analogue of the unmixedness theorem for R - s e q u e n c e s holds.

P R O P O S I T I O N 5.21. Let gr I = n and suppose that I can be generated by

elements. If P e A s s ( R / I m) f o r any m, then gr P =n.

Proof: By P r o p o s i t i o n 5.18 we m a y assume that I = (a I ,..., a n ) with a I ,..., an

an a s y m p t o t i c sequence. If P e A(I,m) t h e n by P r o p o s i t i o n 3.9, P e A (a 1 , . . . , an),

so that a I ,..., an is a maximal asymptotic sequence coming from P. Thus

gr P = n.

W e show two ways asymptotic sequences differ from R-sequences.

If x I , ..., Xn is an R - s e q u e n c e in R and if the images of Xn+ I ,..., x m

are an R - s e q u e n c e in R/(x I ,..., x n), then x I,..., x n , Xn+ 1 ,..., X m is an R-

sequence in R. The analogue for asymptotic sequences fails. Let (R,M) be a

local ring w i t h exactly two m i n i m a l primes p and q, with depth p = 1 and

depth q = 2. Pick a e M- (qUp) and b c M with b not in any prime m i n i m a l

over (q,a). Since a @ qUp, a is an asymptotic sequence in R. Also, the image

of b is easily seen to not be in any minimal prime of R/aR. Thus that image is

an asymptotic sequence in R/aR. Now a,b is not an asymptotic sequence in R,

since depth p= 1 implies z(M) = i, so that gr (M) = i.

W e have p r e v i o u s l y noted that if xI ... x is an R-sequence, and if

' ~ n

P e A (x I ,..., x n) then P e A s s ( R / ( x I ,..., Xn)). The analogue for asymptotic

sequences fails. To see this, as in [N, E x a m p l e 2, pp. 203-205] let (R,M) be a

2-dimensional local domain with R containing exactly two m a x i m a l ideals, N1 and

N 2 , with heights 1 and 2 respectively, such that M=N INN 2 and such that

R=R+Ra~ with Nl=O~. Let ~ e N 2 -N 1 • Note that ~ e NINN2=M. We claim

 41

that ~R=pRNR. Clearly ~---Rc ~ p R n R ~ pRNR. Now if rl+r2~ e R+R~=R with

p(r l + r 2 ~ ) = s e R, then to show s e ~R it is enough to show ~r I e ~pR since

clearly Pr2~ e c~pR. Since p e N 2 , s e M. Also Pr2~ e M. Thus ~r I e M =

N IN N 2 • As p @ N I , rI e N I=~R. Therefore pr I e ~ n R = ¢ i p R as desired. Thus

c~pR = DR N R as claimed. Now a = ~,p is an asymptotic sequence in R, and since

height N I = I implies that z(M) = I so that gr M = I, we have M e A ((a)). How-

ever M @ Ass(R/(a)) since if M= ((a): b) for some b e R, then (a)=o~R =

SRNR would imply M= (~R: b)

NR. By Lemma 1.2 we see that N2 is a prime
R
divisor of ~R. This is impossible since R is a Krull domain and height N 2 =2.

The next proposition is due to Keith Whittington.

PROPOSITION 5.22. The following are equivalent for a Noetherian domain R.

i) R satisfies the altitude formula,

ii) for any finitely generated extension domain T of R, and any height 1

prime Q of ~, height Q N T = i.

Proof: i)~:=> ii) is trivial since a finitely generated extension domain of aNoetherian

domain satisfying the altitude formula also satisfies it.

ii) ~> i): Suppose that i) fails. Then by Proposition 5.11, there is an ideal

I = (a I , ..., an) of the principal class and a prime P e A (I) with heightP>~(Ip).

Note that ~ is in the principal class in Rp so that ~ ( ~ ) =n. By Proposition

3.20, we may assume that T = R [ a l / a n,,.., an_I/a n ] is such that ~ contains a

height 1 prime Q lying over P. By [D, Corollary 2], we see that PT is a prime

of height equal to height P - (n-l) > i. Since PT ~ Q N T , (ii) fails.

 CHAPTER V!: Asymptotic Sequences Over Ideals

The first published result involving asymptotic sequences was by Rees [Rs2].

He defined an asymptotic sequence over an ideal, as follows. (Actually, he called

it an asymptotic prime sequence over I.)

DEFINITION. Let I be an ideal in a Noetherian ring R. Elements x I , .. ., x n of

R are called an asymptotic sequence over I if (l,x I ,..., Xn) # R and if for all

i = l,...,n, x i ~ A (l,x I ,..., Xi_l).

Rees then let r be the length of a maximal asymptotic sequence over I in a

local ring (R,M), and showed that r < height M - ~ ( I ) , and that equality holds

if R is quasi-unmixed.

As we now have more tools at our command than did Rees, we will go a bit

further, showing that in any local ring, all maximal asymptotic sequences over I

have the same length, which we characterize. This was done by Katz in [Kz2]. We

then show that if x I ,..., x n is an asymptotic sequence over I, it is also an

asymptotic sequence modulo I.

We begin with a nice result of Rees, proved in [Rs2, Theorem 2.6].

PROPOSITION 6.1. Let I be an ideal in a local ring (R,M). Let ~ = R [ t - l , lt] be

the Rees ring of R with respect to I, and let x e M with x not in any minimal

prime divisor of t-l~. Then ~(l,x) =~(I) +I.

Proof: Write ~(I) and ~(l,x) for the Rees rings of R with respect to I and

(!,x) respectively. Also let Y be an indeterminate and write ~(I,Y) for the

Rees ring of R[Y] with respect to (I,Y). Let ~: R[Y] + R via Y + x. This

induces an obvious homomorphism ~ of ~(I,Y) onto ~(l,x). We claim that

ker ~ c rad[(t'l,x)~(l,Y)]. However, we will defer the proof of the claim briefly.

Using the above claim, since x e M, we see that ker ~ ~ rad[(t'l,M,Y)~(l,Y)].

Therefore, since ~ maps (t'I,M,Y)~(I,Y) onto (t-l,M)~(l,x), we see that

induces an isomorphism from ~(I,Y)/$(t'I,M,Y)~(I,Y) onto ~(l,x)/J(t-l,M)~(i,x).

Furthermore, $ carries the maximal ideal N = ... + R [ Y ] t - I + (M,Y) + (l,Y)t+ ...

 43

of ~(I,Y) to the maximal ideal N ' = ...+ Rt - I + M + (I,x)t +... of ~(I,x). Since

~(I,x) is the height of N'/(t-l,M)~(I,x), our isomorphism shows that ~(I,x)

equals the height of N/(t-I,M,Y)~(I,Y).

Since Y = t-l(yt), we see that ~(I,Y) =R[Y,t-I,(I,Y) t] =R[t-l,It,Yt] =~(I)[Yt].

Letting Z=Yt, Z is an indeterminate over ~(I) and we have ~(I,Y) =~(I)[Z].

Now N = (t-I,M,Y, (I,Y)t)~(I,Y) when written in terms of Z becomes

(t-I,M, Zt-I, It,Z)~(I)[Z] = (p,Z)~(I)[Z] with p = (t-l,M, It)~(I) =...+ Rt - l + M + I t + ....

Similarly, (t-I,M,Y)~(I,Y) is q~(I)[Z] with q = (t'l,M)~(1) .... Rt - I + M + M I t + ....

Thus ~(I,x) =height N /(t-I,M,Y)~(I,Y) =height (p,Z)~(I)[Z]/q~(I)[Z] =height p/q+l=

~(I) +i.

We now return to proving our earlier claim. First note that ker ~ = (Y-x)R[Y]

shows that ker ~ = E [ ( Y - x ) N (I,y)n]t n, n e Z. Since for m ! 0, (I,Y) m = R [ Y ] , we

see that if a e ker ~, then for large n, ~t -n e (Y-x)~(I,Y). Let ~ be minimal

over (t-l,x)~(I,Y) = (t-l,y-x)~(I,Y) (since Y = t-iZ), and assuming our claim to

be false, choose a e ker ~ - @ . For large n, ~t -n e (Y-x)~(I,Y), and since ~ @

-i
we see that after localizing at ~, that t is in the radical of the ideal gen-

erated by Y-x. Since ~ is minimal over (t-l,y-x)~(I,Y), we now see that

is minimal over (Y-E)~(I,Y). Thus height @ < i. Thinking of ~ as a prime in

~(I)[Z], we must have height ~ N ~ ( I ) < I. As t -I is a regular element in

~N~(I), height @ N ~ ( I ) = i. By the choice of x, x @ ~N~R(I), a contradiction

since x e ~. This completes the proof of the claim, and of the proposition.

COROLLARY 6.2. If Xl ''''' rX is an a s ~ p t o t i c sequence over I in a local ring,

then ~(I,x I ,..., Xr) =~(I) +r.

Proof: This follows from Proposition 6.1 and Proposition 3.10.

DEFINITION. x I ,..., x r is a maximal asymptotic sequence over I if it is an

asymptotic sequence over I but x I ,..., x r, Xr+ 1 is not an asymptotic sequence

over I, for any Xr+ 1 .

We now show that in a local ring, all maximal asymptotic sequences over I

have the same length.

 44

PROPOSITION 6.3. [Kz2] Let I be an ideal in the local ring (R,M). Let

x I ,..., x r be a maximal asymptotic sequence over I. Then

r=min{height(M /q) -~(IR +q/q) I q is a minimal prime in the completion R }.

Proof: It follows easily from Proposition 3.18 that for any minimal prime q of

R , xl+q,...,Xr+ q i s an a s y m p t o t i c sequence over ]21_*+q/q. By L e r ~ a 4 . 3 and

Corollary 6.2, ~(IR~'+q/q)+r=£((I, Xl,...,Xr)R*+q/q) < h e i g h t N*/q. Now by t h e

-4¢
maximality of x 1,..., xr, M e A (I,x 1,..., Xr) and so P r o p o s i t i o n 3 . 1 8 shows

that for some minimal p r i m e q of R , M /q e A ((I, Xl,...,Xr)R +q/q). Since

R*/q satisfies the altitude formula, by Proposition 4.1,

height M * / q = ~((l,Xl,...,Xr)R +q/q) =~(IR*+q/q) +r, and we are done.

COROLLARY 6.4. Let r be the length of a maximal asymptotic sequence over I in

a local ring (R,M). Then

(i) r ! height M -~(I)

(ii) The following are equivalent.

(a) R is quasi-unmixed

(b) Equality holds in (i) for all ideals I

(c) Equality holds in (i) for all ideals I in the principal class.

Proof: (i) By Lemma 4.2, there is a minimal prime q of R with ~(I) = ~ ( I R ) =

~(IR +q/q). By Proposition 6.3, r <_ height(M /q) - ~ ( I R * + q / q ) = h e i g h t ( M /q)- ~(I)

<__ height M - ~(I).

(ii) (a) ~ (b). This is easy using Proposition 6.3, Lemma 4.2, and the fact that

height M = h e i g h t ~ /q) for any minimal prime q in the completion of a quasi-un-

mixed local ring.

(b) ~ (c) is irmmediate.

(c) ~ (a). Let gr M = z(M) = n, and let x I,..., x n be a maximal asymptotic

sequence from M. Let I = (xI , ..., Xn). Now Lermna 5.2 shows that I is in the

principal class. The maximality of our sequence gives M e A (I), so in this

case, r = 0. By (c), height M = ~ ( 1 ) . But ~(I) lies between height I and the
 45

number of generators of I, both of which are n° Thus ~(I) = n . Therefore

height M = n = z(M), showing that R is quasi-unmixed.

W e now wish to show that if x I,.°., x n is an asymptotic sequence over I,

then it is also an asymptotic sequence modulo I (but the converse fails). We

start by showing the numbers compare correctly (Proposition 6.6). First, a lermna.

LEMMA 6.5. Let I be an ideal in a local ring (R,M) and let P be a prime min-

imal over I. Then ~ (I) ~ h e i g h t P.

Proof: Since the minimal number of generators of In is at least as great as the

minimal number of generators of ~ for all n, the Hilbert polynomial character-

ization of analytic spread shows that ~(I) ~ ~ ( ~ ) . Now ~ is Pp-primary, so it

is well known that ~(~) =height P.

PROPOSITION 6.6. Let I be an ideal in a local ring (R,M) and let r be the

length of a maximal asymptotic sequence over I. Then r ~ gr (M/I).

Proof: Since (R/I)* ~ R * /IR * , there is a prime p of R* , minimal over IR * ,

such that gr (M/I) = depth p. Let q be a minimal prime of R with q ~ p.

Since R*/q is quasi-unmixed, gr* (M/I) = depth p = height (M*/q) - height (p/q)

height(M /q) - ~ ( I R +q/q) (by Lemma 6.5) > r (by P r o p o s i t i o n 6.3).

LEMMA 6.7. Let I be an ideal and let x I ,..., x n be an asymptotic sequence over

I and also an asymptotic sequence modulo I. Let P be a prime containing I

such that P/I e A (x l + I , . . . , x n + I ) . Then P e A (I,x I ,..., Xn).

Proof: W e may localize, and assume that R is local at P. The hypothesis shows

that x l+I,.,.,x n+I is a maximal asymptotic sequence from P/I. Thus gr (P/I) = n .

By Proposition 6.6, in the ring (R,P) the length of a maximal asymptotic sequence

over I cannot exceed n. As x I ,..., Xn is an asymptotic sequence over I, it

must he maximal. Thus P e A (I,x I ,..., Xn).

PROPOSITION 6.8. Let x I , ..., x n be an asymptotic sequence over I. Then

x l + I , ...,x n + I is an asymptotic sequence in R/I.

 46

Proof: W e induct on n. For n = I, the hypothesis is that xI is not in any

prime contained in A (I). In particular, xI is not in any prime minimal over I.

Thus xI + I is an asymptotic sequence in R/I. Suppose now that we have

x l+l,...,x i+l an asymptotic sequence in R/I for i < n. Let P/I be prime

in R/I with P/I e A (x I + l , . . . , x i + l ) . Then Lemma 6.7 shows that

P e A (l,x I ,..., xi). Since x I ,..., x i, x i + I is an asymptotic sequence over I,

x i + I ~ P. Thus xi+ I + I @ P/I, and we see that x l+l,...,xi+ I+I is an asymptotic

sequence in R/I.

The converse of Proposition 6.8 is easily seen to fail, since A (I) can

contain primes not minimal over I.

PROPOSITION 6.9. Let I be an ideal in a local ring (R,M), and let Yl ''''' Yn

be an asymptotic sequence from I. Then there is a maximal asymptotic sequence

over I, x I ,..., x r, such that Yl ''''' Y n ' Xl ''''' Xr is an asymptotic sequence.

In particular, r+gr I ~ gr M.

Proof: Let r be the length of a maximal asymptotic sequence over I. If r = 0,

we are done. If r > 0, then M @ A~k (I), and so by Lemma 5.7, M ~ A (Yl ''''' Yn )"

Pick xI e M with x I @ U [P e A (I)} and x I @ U {P c A (Yl ''''' Yn )}" Now xI

is an asymptotic sequence over I, and the length of a maximal asymptotic sequence

over (l,Xl) is r- I (since if x I , x~ ,..., X'r is a maximal asymptotic sequence

over I, then x~ ,..., X'r is a maximal asymptotic sequence over (l,Xl)). Since

the choice of xI assures that Yl ''''' Y n ' Xl is an asymptotic sequence, we may

use induction.

We now determine when the inequality in Proposition 6.9 is equality. First

*
we need another characterization of gr I.

PROPOSITION 6.10. Let I be an ideal in a local ring (R,M). Then gr ( I ) =

min{height I R * + q / q ] q i s a m i n i m a l p r i m e i n R , t h e c o m p l e t i o n o f R}.

Proof: Let x I , °.., x n be a maximal asymptotic sequence from I. By Lemma 5.1,

xl+q,...,Xn+ q is an asymptotic sequence in R*/q for any minimal prime q of R*.

 47

Thus gr (IR +q/q) ~ n. As R /q i s quasi-ur~nixed, C o r o l l a r y 5.8 g i v e s

height IR +q/q ~ n = gr I. Now pick P e A (xI,..., Xn) with I ~ P. By two uses

of Proposition 3.18, there is a P of R with P nR =P and a minimal prime

q ~ P , with P*/q e ~ ( ( X l , . . . , X n ) R +q/q). C o r o l l a r y 5.8 shows t h a t height P*/q=

gr*P*q/= n. As P * FIR=p, IR* + q / q ~ P * /q, so h e i g h t IR * + q / q ~ n = gr * I . As we

already have the other inequality, we are done.

PROPOSITION 6.11. Let I be an ideal in a local ring (R,M), and let r be the

length of a maximal asymptotic sequence over I. The following are equivalent.

(i) r+gr I = gr M.

(ii) For some minimal prime q of R*, ~(IR +q/q) = height(IR +q/q) = gr*l

and r = depth q - h e i g h t (IR + q / q ) .

(iii) The equalities in (ii) hold for every minimal prime q of R satisfying

depth q = gr M.

Proof: (i) ~ (iii): Let q be any minimal prime of R with depth q = gr Mo

We have that depth q = h e i g h t M*/q = ( h e i g h t M * / q - h e i g h t (IR + q / q ) ) + h e i g h t (IR~+q/q)

>__ (height M /q- ~(IR +q/q))+height(IR +q/q), the inequality holding since height

does not exceed a n a l y t i c s p r e a d . Now u s i n g P r o p o s i t i o n s 6.3 and 6.10,

(height M*/q-~(IR*+q/q)) +height(IR*+q/q) >_ r + g r I. Using our choice of q, and

(i), we also have r+gr I = g r M = depth q. As we began and ended with depth q,

we see that e q u a l i t y holds throughout, and ( i i i ) easily follows.

(iii) ~> (ii) is immediate.

(ii) ~ (i). Let q be as in (ii). We have that gr*M <_ depth q = height M*/q =

(height M * / q - h e i g h t ( I R +q/q)) +height(IR +q/q) = r + g r * l <__ gr M, the last two

Steps using (ii) and Proposition 6.9. Thus (i) holds.

COROLLARY 6.12. Let I he an ideal in a local ring (R,M), and let r be the

length of a maximal asymptotic sequence over I. Then r + gr I = height M if and

only if R is quasi-unmixed and height I=~(I).

 48

, , ,
Proof: Assume that r+gr I =helght M. Since r+gr I ~ gr M ~ height M, we have

gr M = height M, which shows that R is quasi-unmixed. By Corollary 6,4,

r +~(I) = height M. The hypothesis and Corollary 5°8 show ~(I) = gr I = height I.

Conversely, if R is quasi-unmixed and height I = ~(I), then Corollaries 6.4 and

5.8 give r+gr l=height M.

Our next lemma is due to Ratliff.

LEMMA 6.13. Let P e A (I), let al,... , an be an asymptotic sequence over I,

and let Q be a prime minimal over (P,a 1 , ..., an). Then Q e A (I,a 1 ,..., an).

Proof: We may localize at Q, and then using proposition 3.18, we may go to the

completion and the work modulo a minimal prime. That is, we may assume that (R,Q)

is a complete domain, which must be quasi-unmixed. Now as mentioned in the proof

of Lemma 6.5, ~(I) ~ ~ ( ~ ) . Using Corollary 6.2 and Proposition 4.1,

~(l,a I ,..., a n ) = ~ ( I ) + n ~ ~(~)+n ~ ~(~)+n= height P + n ~ height P +height Q / P =

height Q since R is eatenary. Using Lermna 4.3, we see that ~(l,a I ,°.., an) =

height Q. Thus Proposition 4.1 gives the result.

In Lermma 6.13 it is not enough to assume that Q e A (P,a I ,..., an). To see

this, consider the example and remark preceeding Lemma 4.5. That example gave a

3-dimensional quasi-unmixed local ring (R,M) and a height I prime P with

~(P)=2, such that A (P)={P}. Pick O~x e P and let I=ER. Then ~(I)=I.

By Proposition 4.1, M @ ~'=(I). Let a be an asymptotic sequence over I. Thus

-.@
a ~ P. Since A (P)= [P}, a is also an asymptotic sequence over P. By Corollary

6.2, ~(l,a) = ~ ( I ) + i = 2 while ~(P,a) =~(P) + I = 3. By Proposition 4.1,

M e A (P,a) but M @ A (l,a).

PROPOSITION 6.14. Let I be an ideal in a local ring (R,M), and let r be the

length of a maximal asymptotic sequence over I. Then r <_ min{little depth P I

P e A (I)} (little depth P is the length of a shortest saturated chain of primes

between P and M.)

 49

Proof: Let P e A (I) and suppose that n=little depth P is minimal among

primes in A (I). Consider a chain of primes P = P 0 c P1 c . . . ~ Pn-1 c Pn=M" We

induct on n. If n = 0, M = P e A (I) and so r = O. If n = i, then by definition

~k
of n, M ~ A ( I ) and we p i c k a e M t o be an a s y m p t o t i c s e q u e n c e o v e r I . Since

a ~ P and since p c M is saturated, M is minimal over (P,a). By Lemma 6.13,

M c A (!,a) showing that r = 1. Now a s s u m e n > 1. By [M2, Theorem 5]

Pn-I
we can assume that height -~--0 = n-l. Now the definition of n, and the fact that

n > i, show that neither Pn-I nor M=P are in A (I). Thus
n
Pn-I ~ U[Q e A*(1)}. Pick a e Pn-I with a an asymptotic sequence over I. Of

course a ~ P. Let (P,a) c Pi c Pn-I with P~ a prime minimal over (P,a).

p' P
I n-i
Thus height T0 I, and since height Pn--~--0= n-I we must have height ~ < n-2.
PI --
Thus since Pn-I c M is saturated, we see that little depth Pi <- n-1. By Lemma

6.13, we see that Pie A (i,a). As the length of a maximal asymptotic sequence

over (l,a) is r-I, by induction we have r-i ~ n-l. Thus r ~ n.

In a quasi-unmixed local ring, we can characterize when the inequality of

Proposition 6.14 is equality. Of course since R is catenary, little depth

equals depth.

PROPOSITION 6.15. Let I be an ideal in a quasi-unmixed local ring (R,M) and

let r be the length of a maxi~nal asymptotic sequence over I. The following are

equivalent.

i) r=mln[depth P I P c A (I)]

ii) ~(I) =max[height PIP e A*(1)}

iii) I(1) = ~ ( ~ ) for some P e A (I).

Proof: By Proposition 6.4, r + ~(I) = height M. Since R is catenary, depth P +

height P = height M for all primes P. The equivalence of i) and ii) is now

straightforward, ii) ~ iii) is immediate from Proposition 4.1. Finally, suppose

~(I)=~(~) with P e A (I). By Proposition 4.1, 8(1) =height P. The truth of

ii) now follows from the next lemma.

 50

LEMMA 6.16. Let I be an ideal in a quasi-unmixed local ring. Then (I) >

height P for all P e A (I).

Proof: ~(I) >__ ~ ( ~ ) as we have already seen. N o w use Proposition 4.1.

PROPOSITION 6.17. Let I be an ideal in a local ring ~,M) and let x I ,..., x n

be an asymptotic sequence over I. Then any permutation of x I,..., x n is also an

asymptotic sequence over I.

Proof: Since x I , ..., x n is an asymptotic sequence over I if and only if

xl+q,...,Xn+ q is an asymptotic sequence over IR +q/q for any minimal prime q

of R (Proposition 3.18), we may assume that R is quasi-unmixed. Also note that

x I,..., x n is an asymptotic sequence over I if and only if x I ,..., x.l is an

asymptotic sequence over I, and x i + I , ..., x n is an asymptotic sequence over

(l,x I ,..., xi). This, together with the fact that any permutation is a product of

adjacent transpositions, easily allows us to reduce to the following. We assume

that x,y is an asymptotic sequence over I, and must show that y,x is an asymp-

totic sequence over Io

Suppose that P e A (I) and y e P. Let Q be a prime minimal over (P,E).

B y Lemma 6.13, Q e A (l,x). As y e P c_ Q, our assumption on x,y is contradicted.

Therefore we see that y is an asymptotic sequence over I.

N o w suppose p e A (l,y) and x e p. W e m a y localize at p, maintaining all

of our hypotheses. B y Proposition 4.1 and Corollary 6.2, height p = ~ ( l , y ) =~(I) +I.

However, L e m m a 4.3 shows that height p > ~(l,x,y) = ~ ( I ) + 2 . This contradiction

completes the proof.

LEMMA 6.18. Let I be an ideal in a local ring R, and let I' be the image of

I in a h o m o m o r p h i c image of R. Then ~(I') ! ~(I).

Proof: For all n, I n requires at least as many generators as I 'n. Thus the

lerm~a follows from the Hilbert polynomial characterization of analytic spread.

LEMMA 6.19. Let I be an ideal in a locally quasi-unmixed Noetherian ring R.

Let Yl ''''' Y m be elements of R which form both an asymptotic sequence and an

 51

asymptotic sequence over I. Let primes denote modulo (Yl ''''' Ym )° If

e' e ~ ( I ' ) , then P e ~=(l,y I,..., y m ).

Proof: W e localize at P. Lemma 5.2 implies that R' is quasi-unmixed. P r o p o s i t i o n

4.1 says height P' = ~ ( I ' ) . N o w using the principal ideal theorem, Lemma 6.18, and

Corollary 6.2, we have height P ~ height P ' + m = ~ ( l ' ) +m ~ ~(1)+m=~(l,Yl,.~°, ym).

Finally L e m m a 4.3 shows height P = ~ (I,y I , ..., y m ) and so P r o p o s i t i o n 4.1 gives the

result.

PROPOSITION 6.20. Let I be an ideal in a local ring (R,M), and let X l , . . . , xn

be an asymptotic sequence over I. Then x I ,..., Xn is an asymptotic sequence.

Proof: By Proposition 5.1, it is enough to show that the images of x I,..., xn

*/q *
form an asymptotic sequence in R for each minimal prime q of R o As Propo-

sition 3.18 implies that those images are an asymptotic sequence over IR*+q/q,

we may return to (R,M) with the added assumption that it is a quasi-unmixed local

ring.

Suppose xI is in q, a minimal prime of R. Choose P minimal over I + q.

By Corollary 3.13, P e A (I). As x I e P, we have a contradiction. Thus xI is

not in any minimal prime, and so is an asymptotic sequence. N o w inductively assume

that x I ,~.., Xn_ I is an asymptotic sequence. L e t t i n g primes denote modulo

(x I,..., Xn_l) , we claim that x n' is an asymptotic sequence over I' This is

ir~nediate, using L e m m a 6.19. Theorefore, the case n = I shows that x' is an

n
asymptotic sequence in R'. Thus xn is not in any prime minimal over

(x I ,..°, Xn_l). B y Lem~a 5.3, height(x I ,..., Xn_ I) = n - I, so that

height (x I ,..., Xn) = n . Lemma 5.3 now gives the result.

We wish to show that if x I ,.°., x n is an asymptotic sequence over I in a

local ring, then ~(I) = ~ ( ( I , x I ,..., Xn)/(x I ,..., Xn)). W e begin with a lermna due

to Brodmann, whose hypothesis will be significant in the next chapter.

LEMMA6.21. Let I be an ideal in a local ring (R,M) and suppose the images of
 52

Xl~'''~ ~ modulo ITM are an R-sequence in R/I m for all large m. Then

~(I) =~((I,x 1,..., Xn)/(x I,..., Xn)).

Proof: We induct on n. For n = i, we have x I ~ U{P e A (I)}. We claim that for

large n, Im/Ml TM ~ l'm/M'l 'm, the prime denoting modulo x I • To see this, note

that if y e Im and y' e M' I 'm, then y = z+rx I with z ~ MI m. As rx l = y - ze ITM

the hypothesis on xI shows that r e ITM, so that rx e MI TM- Thus y e MI TM. Now

this isomorphism shows that ITM and I 'm require an equal number of generators.

Thus ~(I) =~(I'), proving the case n = I.

We inductively assume the result for n - I, and prove it for n. It is straight

forward to verify that R/I m /x I (R/I TM) ~ R'/I 'm. Since the images of x 2 , • . ., x n are

an R-sequence on the left hand side, going to the right hand side and applying the

case n-l, we have ~(l')=~((l',x2, .., x )/(x ... Xn))=~((l,Xl,..., Xn)/

(Xl,... , Xn)). Since ~(I) =~(I'), we are done.

PROPOSITION 6.22. Let I he an ideal in a local ring (R,M), and let x ! ,..., x n

be an asymptotic sequence over I. Then ~(I) =~((l,Xl,..., Xn)/(Xl,.-., Xn))-

Proof: Suppose n = I. Assume momentarily that R is a complete local domain, hence

analytically unramified. Thus for large m, J = ITM is normal (that is, all powers

are integrally closed) [SO]. We easily see that ~(I) =~(J) and A (I) (J) =

A (J). Since x I 4 U[P e A (I)}, by Lermma 6.21 we have ~(J) =~((J,Xl)/(Xl) ). Now

J = ITM implies ((l,Xl)/(Xl))m ~ (J,Xl)/(Xl) c_ ((l, Xl)/(Xl)) TM, from which we see

that ~((l, Xl)/(Xl))=~((J, X l ) / ( X l ) ) = ~ ( J ) = ~ ( 1 ) , completing this case. Now assume

n =l and (R,M) is arbitrary. Since ~(I) = ~ ( I R ) and ~((l, Xl)/(Xl)) =

~((l, Xl)R /XlR ), we may suppose that R is complete. By Lemma 4.2, for some min-

imal prime q, ~ ( I ) = ~(l+q/q). In the complete local domain R/q, we get ~(I) =

~(l+q/q) =~((l,q,xl)/(q,xl)) < ~((l,Xl)/(Xl)) <__ ~(!), by two uses of Len~na 6.18.

This finishes the case n = i.

Now inductively assume the result holds for Xl,..., Xn_ I • We first treat the

case that (R,M) is quasi-unmixed. In this case, Lemma 6.19 and Proposition 6.20
 53

show that x' is an asymptotic sequence over I', the primes denoting modulo
n
(x I ,..., Xn_l). By the case n = i, ~(I') = ~ ( ( I ' , x ~ ) / ( X n ) ' ) = ~ ( ( I , x I ,..., Xn)/

(x I ,..., Xn)). Since our induction shows that ~(I) = ~(I'), we have proved our

result for R quasi-unmixed.

In general, we have ~(I) = ~ ( I R +q/q) for an appropriate minimal prime q

of R . Since x l+q,...,x n+q is an asymptotic sequence over IR +q/q, and

since R*/q is quasi-unmixed, the preceding shows that ~(I) = ~ ( ( I , x I ,..., x n, q)R*/

(x I ,''', x n, q)R*) ! ~((I,x I ,''', Xn)R * /(x I ,''', Xn)R * ) = ~ ( ( I , x I ,''', Xn)/

(x I ,''', Xn)) ! ~(I), using Lemma 6.18. This completes the proof.

COROLLARY 6.23. Let I be an ideal in a locally quasi-unmixed Noetherian ring.

Let Yl ''''' Ym be an asymptotic sequence over I, and let primes denote modulo

(Yl ''''' Ym )" Then P' e ~ ( I ' ) if and only if P e A (I,y I ,..., ym ).

Proof: We may localize at P. Since Proposition 6.20 tells us that Yl ''''' Ym

is an asymptotic sequence, we use Lemma 6.19 for one direction. Thus suppose that

P e A (I,y I ,..., ym ). By Proposition 4.1, height P = ~(I,y I , ..., ym ) = ~(I) +m,

using Corollary 6.2. Now Lemma 5.2 shows that height P' = h e i g h t P -m=~(I) = ~(I'),
--4.~
the last equality by Proposition 6.22. Thus Proposition 4.1 gives P' e A (I').

When studying A (I) it obviously does no harm to replace I by any power of

I, or by any ideal having the same integral closure as I. Therefore if as defined

below, I ~ J, then A (I) = A (J).

DEFINITION. Let I and J be ideals in a Noetherian ring. I and J are pro-

jectively equivalent (I~J) means for some positive integers n and m, i n = j m

Projective equivalence was studied by Samuel in [Sm} and in Chapter I0 we

will have an interesting use for projective equivalence. For now, we present the

following.

PROPOSITION 6.24. Let I and J be projectively equivalent. If x I , ~ .., x n is

an asymptotic sequence over I, then it is an asymptotic sequence over J.

 54

Proof: As A (I)=A (J), the statement is true for xI . Assume it true for

x I ,..., Xn_ 1 . To show it true for x I,..., x n, it suffices to show that

A (I,x I ,..., Xn_l) = A (J,x I ,..., Xn_l). Suppose P c A (I,x I,..., Xn.l). We may

assume that R is local at P, and for an appropriately chosen minimal prime q

* p*/q --* , *,
of R , we have e A ((I, Xl, ..., Xn_l)R +q/q). As R /q is quasi-unmixed,

height P / q = ~ ( ( I , x I,..., Xn_l)R +q/q) = ~ ( I R * + q / q ) +n-l, using Corollary 6.2.

Now since i m = jk for some positive integers m and k, it is straightforward

to see that ~(IR * +q/q) = ~ ( J R * + q / q ) . Thus height P */q = ~ ( J R * + q / q ) + n . 1 =

~((J,x I ,..., Xn.l)R +q/q), since our inductive assumption shows that

x I + q, ...,Xn_ 1 + q is an asymptotic sequence over JR +q/q. Therefore

p*/q -~" , -,
c A ((J,x I,°.°, Xn.l)R +q/q) and so P e A (J,x I ,..., Xn_l)° The reverse

is identical.
 CHAPTER VII: Asymptotic Grade

In [B2] Brodmann shows that if I is an ideal in a local ring ~,M), then

the sequence grade(R/I n ) n = 1,2,... eventually becomes constant, and if this con-

stant is ~, then ~(I) ! height M - ~ . This chapter presents these proofs and

tries (with only limited success) to extend them to grade(R/In). (Earlier work of

Burch [Bu] has shows that ~(I) ~ height M - lim inf grade(R/In), at least when

R/M is infinite.)

Our first proposition will be an extension of Proposition 1.3 to modules, and

as the proof is essentially identical, it will be omitted.

PROPOSITION 7.1. Let T = ER n , n ~ 0, be a Noetherian homogeneous graded ring,

and let A = ZA n , n ~ 0, be a finitely generated graded T-module. Then for large

n, ASSR0(An) = ASSR0(An+l) .....

The next result is due to Brodmann, generalized somewhat by D. Katz in order to

glean some information about grade(R/In).

PROPOSITION 7.2. Let R be a Noetherian ring, and let A be a finitely generated

R module. Let A = A 0 ~ A I ~ A 2 ~ ... be a sequence of submodules, and suppose

that there is an ideal I of R and an integer h > 0 such that for all n > 0

i n - h A ~_ An --DAn+ 1 ~-- I.An • (Here, I n-h= R when n<h.) Then

(i) ASSR(An_I/An) stabilizes for large n.

(ii) ASSR(A/An) stabilizes for large n.

(iii) If R is local, then gradeR(A/An) stabilizes for large n, and this

stable value is equal to or less than gradeRA.

(iv) If R is a local ring and if n is large enough that gradeR(A/An)

has stabilized, say at ~, then if x I , ..., x~ is an R-sequence on

A/A n, it is also an R-sequence on A/~ for all k h n.

Proof: With X an indeterminate, let T=R+IX+I2X 2 +... , and let

B = A 0 + A I X + A 2 X 2 +... , (in both cases meaning finite polynomials). B is a

T-module. The hypothesis shows that A h + A h + I X +Ah+2X2 +... is contained in

A+IAX+I2AX 2 + .... This last is a graded T-module, finitely generated by the

 56

same elements which generate A over R° Since T=R[IX] is Noetherian,

Ah+Ah+IX+Ah+2X2 +... is a finitely generated T-module. We now easily see that

B is a finitely generated T-module. Since AI+A2X+A3X2 +... is a submodule of

B, we have that A/A I + A I / A 2 X + A 2 / A 3 X 3 +... is a finitely generated T-module.

Using Proposition 7.1, since T0=R , we have (i).

For (ii), consider the exact sequence 0 + An_I/A n + A/A n + A/An_ 1 ~" O, which

shows that ASSR(A/An) ~ ASSR(A/An_I)UASSR(An_I/An). Since for large n,

ASSR(An.i/An) =ASSR(An_2/An_I) ¢~ ASSR(A/An.I) , we have ASSR(A/An) ~ ASSR(A/An_I)-

As these sets are finite, they must stabilize, proving (ii).

For (iii), let (R,M) be a local ring, and suppose that m is large enough

that ASSR(A/Am) has stabilized. First note that if x e M-U [Q e Ass(A/Am)}

then x is not a zero divisor on A. To see this, suppose that xa = Oo The choice

of x and m shows that a e ~ for all k >__m. However, the hypothesis shows

that N ~ c_ N Ik'hA= 0 for k = 1,2,..., so that a= 0 as desired.

We now induct on grade~A. If gradeRA = O, the preceding paragraph shows that

M e Ass (A/Ak) for all k > m, so that gradeR(A/Ak) stabilizes at O. Now

suppose that gradeRA > 0. If M e ASSR(A/Am), then again grade(A/A k) stabilizes

at O. Now say xeM-U[QcASSR(A/Am)}° By the preceding paragraph, x is not a

zero divisor on A, and so letting ~ =~ +xA/xA, we have gradeRA' = gradeRA-l-

Also the sequence A ' -- A 0' _ A i ~_... satisfies the hypothesis of the theorem, and

so by induction, gradeR(A'/~) stabilizes to a value not exceeding gradeRA'=

gradeRA - I. However, A'/~ ~ A/Ak+XA = (A/Ak)/X(A/Ak). Since x is not a zero

divisor on A/~ , k >__m, gradeR(A'/~) = gradeR(A/~) - I. Thus gradeR(A/Ak) stabi-

lizes to a value not exceeding gradeRA.

For (iv), we induct on ~, the case ~= 0 being trivial. Now suppose that

grade R(A/Ak) = ~ > 0 for all k >_ n, and let x I ,..., x~ be an R-sequence on

A/A n . Letting ~ = ~ +xIA/XlA, we have A'/~ ~ (A/Ak)/x I ( A / ~ ) so that for

k >_ n, gradeR(A'/~) = gradeR(A/~) - i = ~ - i. Thus gradeR(A'/~) has stabilized

for k >_ n, and since our isomorphism shows that x 2 , ..., x~ is an R-sequence

on A'/An, by induction it is also an R-sequence on A'/A~ for k >__n. Again

using our isomorphism, we get (iv).

 57

Remark: The proof of (iii) above is a bit misleading. It does not show that

gradeR(A/An) stabilizes as soon as ASSR(A/An) stabilizes.

Question: concerning (iv) above, is there some n such that if Xl,..., x~ is an

R-sequence on A/A k for some k ~ n, then it is also an R-sequence on A/An?

COROLLARY 7.3. (Brodmann [B2]) Let I be an ideal in a local ring (R,M). Then

grade(R/I n ) stabilizes to a value not exceeding grade R~

Proof: Immediate, letting A n = In , h = O.

As mentioned previously, we do not know if this corollary holds with In re-

placed by In . However, in two instances it does.

COROLLARY 7.4. (Katz) Let (R,M) be an analytically unramified local ring, and

let I be an ideal. Then grade(R/I n ) stabilizes for large n.

Proof: Let A=R and for n > I, A = In . As R is analytically unramified, there

- n

is an h with In'h ~ In for all n > h. Use Proposition 7.2.

COROLLARY 7.5. (Katz) Let (R,M) be a complete local ring, and let I be an ideal.

Then grade(R/I n ) stabilizes.

Proof: Let Z = rad R. Thus R/Z is analytically unramified. Letting A = R/Z and

for n ~ i, A n = In/Z (noting that Z ~ In), and considering the ideal I +Z/Z of

R/Z, there is an h ~ 0 with ( I + Z / Z ) n-h ~ ( I + Z / Z ) n ~2 A n ~ An+ I ~ ( I + Z / Z ) A n •

By Proposition 7.2, gradeR/z(A/An) stabilizes. Thus gradeR(R/I n) stabilizes.

Question: If I is an ideal in a local ring (R,M), will grade(R/I n ) always

stabilize?

By Proposition 7.2 (iv), we see that it makes sense to discuss elements

of R whose images in R/I n form an R-sequence for all large n. We

Xl ~ •.., Xs
now show that such elements form an R-sequence, an asymptotic sequence, and if R

is locally quasi-unmixed, an asymptotic sequence over I.

 58

PROPOSITION 7.6. Let (R,M) be a local ring, and let I he an ideal. Let

Xl,''', X s be elements of R whose images in R/I n are an R-sequence for all large

n. Then x I ,..., Xs is an R-sequence.

Proof: We induct. Suppose xlY = 0. Then xlY e I n for all large n, and by the

nature of x I , y e N In= 0. Thus xI is an R-sequence. Letting primes denote

modulo X l , since R'/I 'n m R/(In,Xl ) ~ (R/In)/Xl(R/In), we see that the images of

x 2,, .', . X's in R'/I 'n is an R-sequence. By induction, x~ ,..., X's

in R', so that x I ,..., Xs is an R-sequence in R.

Note that since the xq,

I ..., x s of Lemma 7.6 are an R-sequence, Lermma 5.13

shows they are an asymptotic sequence.

PROPOSITION 7.7. Let I be an ideal in a locally quasi-unmixed Noetherian ring,

and suppose that the images of x I ,..., x s are an R-sequence in R/I n for all

large n. Then x I ,..., Xs are an asymptotic sequence over I.

Proof: Since xI is not a zero divisor on R/I n for all large n, we see that

x I 4 U [Q e A*(1)}. As A (I) ~ A (I), we have xI an asymptotic sequence over I.

We now inductively assume that x I ,..., Xs. 1 is an asymptotic sequence over I,

..@
and we must show that Xs @ U {P e A (I,x I,..., Xs.l)}. Suppose to the contrary that

xs e P e A (I,x I ,..°, Xs_l). W e may localize at P, and invoke Proposition 4,1

and Corollary 6.2 to see that height P = ~ ( I ) + s - i. Letting a prime denote

modulo (x I ,..., Xs_l) , Lermna 6.21 gives ~(I') = ~ ( I ) = h e i g h t P - (s-l). Since (R,P)

is quasi-unmixed, and since P r o p o s i t i o n 7.6 shows that height(x I ,..., Xs_ I) = s - I,

we have height P' = h e i g h t P - (s-l)=~(I'). By Proposition 4.1, P' e A (I'). (We

could have simply invoked Corollary 6.23, but that uses Proposition 6.22, which

really is not needed.) Since A (I') c A (I'), for large n we have

P' e Ass(R'/I'n), so that P e Ass(R/(In, Xl ,..., Xs_l)). Since x s e P, we have

contradicted that x I,..., x s is an R-sequence modulo I n for large n.

Question: If the images of x I ,..., x s are an R-sequence in R/I n for all large

n, is x I ,..., x s an asymptotic sequence over I even when R is not locally

quasi-unmixed?
 59

Question: Let (R,M) be a local ring and let I be an ideal. Let r be the

length of a maximal asymptotic sequence over I, and let ~ be the eventual stable

value of grade(R/In). Is ~ < r. (Yes, if R is quasi-unmixed, by Proposition 7.7.)

Remark: If the answer to this last question is yes, then our next result would

follow from Corollary 6.4. The proof we give is due to Brodmann.

PROPOSITION 7.8. Let I be an ideal in a local ring (R,M). Let ~ be the

eventual stable value of grade(R/In). Then ~ ~ height M - ~ ( 1 ) .

Proof: We induct on ~, the case ~= 0 following from Lemma 4.3. For ~ > O,

we have M 4 A*(1). Pick x e M-U {Q e A (I)}, and let a prime denote modulo x.

By Ler~na 6.21 we have ~(I) =~(I'). Also, Lermna 7.6 shows that x is not a zero

divisor in R, so that height M' =height M - I . Finally, R'/I 'n ~ R/(In, x) =

(R/In)/x(R/l n) so that ~' = grade R'/I 'n= grade R/I n - I = ~ - I. By induction,

~' ~ height M ' - ~ ( I ' ) and therefore ~ ~ height M - ~ ( 1 ) .

In Proposition 7.8, equality need not occur. Let (R,M) be a 2-dimensional

local domain which is not Cohen-Macaulay and let I be principal. Then ~(I) = i

and ~ = O.

Let (R,M) be a local domain with R/M infinite, and let I be an ideal.

In [Bu], Burch shows that ~(I) ~ height M - lin inf grade(R/In). Propositions 7.2

and 7.8 give a shorter proof, and also show that the lin inf is in fact the

limit ~. However, [Bu] also shows that ~(I) ~ height M - lin inf grade(R/In). As

stated earlier, we do not know if grade(R/I n ) always stabilizes. When it does,

(such as in Corollaries 7.4 and 7.5), we can combine that fact with Burch's second

inequality to get the following.

PROPOSITION 7.9. Let I be an ideal in a local ring (R,M) with R/M infinite,

and suppose that grade(R/I n ) = ~ for all large n. Then ~ height M - ~ ( 1 ) .

In Proposition 7.9, equality need not occur. Let (R,M) be a 3-dimensional

normal Noetherian domain which is not Cohen-Macaulay. Let I =aR# 0 be principal.

As R is normal, I n = (an ) = (an). Thus grade(R/I n ) # 2 or else we would have

 60

grade R = 3 making R Cohen-Macaulay. If for any n, grade(R/I n ) = 0, then

M e A(l,n) ~ ( I ) showing that grade(R/V)= 0 for all large k, and ~ exists

and equals 0. Otherwise, we have grade(R/I n) is neither 2 nor 0, and so in

this case ~ exists and is I. Thus ~ is either 0 or I, while ~(I) = i and

height M = 3.

Question: With ~ as in Proposition 7.8 and ~ as in Proposition 7.9, is ~ ~ ~?

(Note that ~= 0 implies M e ~(I) ~ A*(1) giving ~ = 0.)

While we do not know if the ~ of Proposition 7.9 always exists, we may go to

the completion, and use Corollary 7.5. Since height M = height M and ~(i) = ~(IR )

we have the following.

PROPOSITION 7.10. Let (R,M)be a local ring and let I be an ideal. Let
--#¢
= grade(R*/InR *) for all large n. Then $ ~ height M - ~(I).

Question: Note that grade R/I n = grade R*/InR *. How does grade R/I n compare to

R* / InR * ?
 CHAPTER VIII: When A = A

In Corollary 4.7, we saw that in a 2-dimensional normal N o e t h e r i a n domain,

A (I) = A (I) for any ideal I. In this chapter, we will attempt to identify all

domains having that property, and will almost succeed. In particular, among domains

satisfying the altitude formula, we will see that Corollary 4.7 tells the whole

story.

The path we will take generally follows that laid out by R a t l i f f in [R6]. How-

ever, we will take some shortcuts by using L e m m a 8.3.

Notation: Let @ denote the class of Noetherian domains in which A*(1) = ~ ( I )

for every ideal I.

W e note that the N o e t h e r i a n domain R is in ~ if and only if Rp e @ for

every prime P of R. Therefore we will content ourselves with investigating local

domains in @.

We also note that if J c I c ~ are ideals, then for all n, jn reduces In ,

so that jn c I n c j n Thus A (I) = A (J).

LEMMA 8.1. Let I be an ideal in a N o e t h e r i a n ring R. Let x e I beta regular

element. Then for large n, (In+h : I h) = I n , for all h.

Proof: Consider t as in Lemma l.l(b). Now x(I n+l : x) = I n + I N (x), and by the

Artin-Rees lemma, there is an integer r > 0 such that for large n,

I n + I N (x) = I n+l'r(I r N (x)) <_ xl n+l'r. Thus (In+l : I) ~ (I n+l : x) ~ in+l-r c I t ,

since x is regular. By Lemma l.l(b), for large n we now have (In+l : I ) = I n ,

and so we easily see that (In+h : I h) = I n for all h.

Notation: Let I be a regular ideal in a Noetherian ring. Since (I 2 : I) c

(13 : 12) ~ (14:13 ) ~... eventually stabilizes, we let ~ = (In+l : I n ) for all

large n.

was introduced by Ratliff and Rush [RR].

L~MA 8.2. Let I be a regular ideal in a Noetherian ring.

(i) ~m=l m for all large m.

 62

(ii) m=l m for all large m.

(iii) I=I.

(iv) Let J be an ideal containing I. The following are equivalent.

(a) JC~.

(b) ~=~.
(c) For all large m, ITM = jm

(d) For some m > I, im = j m

(v) Let J be any ideal and suppose that for some m, jm = ITM and jm+l = im+l.

Then J c ~.

(vi) I c I c I.

Proof: ( i ) : For large m, ~ = (Im+l : Im), so ~m c (Im 2+m : Im2 ) = Im, by Lermna 8. I.

As I ~ ~, (i) is clear.

(ii): Using Lemma 8.1, for large m and any h, (Im(h+l) : Imh) = Im. Thus

ira= iTM.

(iii) : For large m, I = (~m+l : ~m). Using (i), this equals (Im+l : ITM) = ~.

(iv): (a) ~ (c): Since I c J c ~, (i) shows that Im=j m for all large m.

(c) ~ (b): For large m, ~ = (Im+l : iTM) = (jm+l : jm) = ~. (b) =3 (d) : Combine (b)
m-I _m m
with (i). (d) ~ (a): Suppose im = j m Then Jl ~_J = I , so J c (ITM : Im-l) c~.

(v) jlm=jjm=jm+l=l m+l. Thus J c (Ira+l: Im) ~ . (Note: Unlike in (iv)d,

Im=j m is not enough. Let R=K[X,Y] modulo (X2 - Y 4 ) . Let I = (X) and j = (y2)..

Then j 2 = 12 but J ~_ ~ = I.)

(vi): Obviously I ~ ~. Now if x E ~, then (i) shows x m ~ im for large m.

Hence x e ~.

LEMMA 8.3. Let 0R,M) be a local ring. Let I c J be ideals with J ~ I.

Then M e Ass OR/(I+MJ) n) for all n > I. In particular, M e A* (I +MJ).

 63

Proof: By Lemma 8•2(iv), for all n ~ i we have in # jn. Thus by Nakayama's

lemma, jn ~ in + M j n Since clearly (I +MJ) n ~ In + M J n, we have jn ~ (I +MJ) n.

However, Mnj n ~ (I +MJ) n, showing that Mn consists of zero divisors modulo

(I+MJ) n. As M is maximal, M ~ Ass(R/(I+Mj)n).

LEMMA 8.4. Let (R,M) be a local domain which is not normal. Then for some

O#a e R, M e A*((a) +M(7)).

Proof: Pick ~ e R-R and O# a e R with ao. e R. Then a~ e aRAR=a--R but

ac~ @ aR. Thus (a) # (a). However we clearly have (a) = (a). Therefore (a) ~_ (a)

and we use Lemma 8.3.

LE~A 8.5. Let a,b be a pair of analytically independent elements in the local

domain (R,M) • Then M e A* ((a2,b 2) +M(a2,b2)).

Proof: Let I = (a,b) and J = (a2,b 2) c 12. Since a,b are analytically independ-

ent, the minimal number of generators of 12n, for n ~ I, is 2n + i. As a2 , b 2

are also analytically independent, the minimal number of generators of jn is

n + i. Hence jn # 12n for any n, and so by Lemma 8.2(iv) 12 ~ 7. However, since

ab satisfies X 2 - a2b 2, we have J ~ 12 ~ . Thus J ~ J, and so we use Lemma8.3.

LEMMA 8•6. Let R be a Noetherian domain with R e @. Let a,b be elements of

R such that no height i prime of R contains (a,b). Then if J = (a2, b2),

A (J)= {P prime I (a,b) ~ P}. In particular, only finitely many primes of R con-

rain (a, b) .

Proof: One inclusion is obvious. Thus suppose (a,b) ~ P . As no height I prime

of ~ contains (a,b), a and b are analytically independent in ~ . By

Lemma 8.5, Pp e A (Jp+Pp~p). Thus P c A (J+P~)=A (J+PJ)=A (J), using that

R e @ and that J c J+pJ c J.

PROPOSITION 8.7• If R e if, then dim R < 3.

Proof: If dim R > 3, there is a height 2 prime P with depth P > I. Pick

a,b in P with no height 1 prime containing both a and b. As depth P > I,

 64

infinitely many primes contain P, and hence contain (a,b). This contradicts

Len~na 8.6.

We now proceed in two cases, first assuming that (R,M) is quasi-unmixed.

PROPOSITION 8.8. Let the local domain (R,M) be quasi-unmixed.* Then R e@ if

and only if either R is 1-dimensional or R is 2-dimensional and normal.

Proof: If dim R = I, then for any I~ 0, A (I) = [M} = A (I). If dim R = 2 and

R is normal, use Corollary 4.7.

Conversely, suppose that R e @ and dim R > I° We will show that R is

2-dimensional and normal. Since dim R > I, choose a,b analytically independent

in R. By Lermma 8.5, M e A*((a2,b 2) +M(a2,b2)). Since R e @ and since

(a2,b 2) ~ (a2,b 2) +M(a2,b 2) ~ (a2,b2), we see that M e ~(a2,b2)° By Proposition

4.1, height M = ~ ( a 2 , b 2) = 2. Thus dim R = 2. Now suppose that R is not normal.

By Le~ma 8.4, for some a e R we have M e A ((a) + M ( ~ ) ) = ~ ( ( a ) + M ( ~ ) ) = ~ ( ( a ) ) -

By Proposition 4.1, height M = Z ( ( a ) ) = i, a contradiction. Thus R is normal.

PROPOSITION 8.9. Let ~,M) be a local domain which is not quasi-unmixed (so

dim R > I).

(i) If dim R = 2, R e

(ii) If dim R = 3 and if R@R, then R e if and only if R contains a

height i maximal, and for all height 2 primes P of R, either ~ is normal

or ~ contains a height i prime lying over P.

Proof: (i) Since the only way a 2-dimensional local domain can fail to be quasi-

unmixed is for to contain a height i maximal, by Proposition 3.19 we see that

M e A (I) for all I#0. As A (I) c A (I), we now easily see that A (1)=A (I).

(ii) Suppose that R e @ Since R ~ R , Lemma 8.4 says M ~ A*((a) +M(a)) =

-_, -.,
A ((a) +M(a)) = A ((a)) for some a e R. By Proposition 3.19, ~ contains a

height I maximal. Now let P have height 2 in R° Of course ~ e @ If

is quasi-unmixed, Proposition 8.8 shows that it is normal. On the other hand,

 85

if ~ is not quasi-unmixed, then as mentioned in (i) above, ~ must contain a

height I prime lying over Pp , so that R contains a height I prime lying

over P.

For the converse, assume that R has the given properties. We will show that

R e @. Now Proposition 3.19 shows that M e A (I) ~ A (I) for all I # 0. Also

a height 1 prime i s i n A (I) i f and o n l y i f i t i s minimal o v e r I i f and o n l y

if it is in A (I). Therefore we need only consider a height 2 prime P and

show P e A (I) if and only if P e A (I). If Rp is normal, then it is Cohen-

Macaulay, h e n c e q u a s i - u n m i x e d , and by P r o p o s i t i o n 8 . 8 , Pp e A (Ip) i f and o n l y

if Pp e A ( ~ ) . If ~ is not normal, then by hypothesis, ~ contains a height I

prime l y i n g o v e r P. By P r o p o s i t i o n 3.19, P e A (I) c A ( I ) for all I # 0. This

completes the proof.

One case eludes our efforts. If (R,M) is a 3-dimensional normal local domain

which is not quasi-unmixed, then we do not know if R must be, might be, or cannot

be in @. (Only recently has it been shown that normal Noetherian domains which

are not quasi-unmixed exist. See [0] or [H2].) The next proposition might help

resolve the issue.

PROPOSITION 8.10. Let (R,M) be a 3-dimensional normal local domain which is not

quasi-unmixed Assume that R/M is infinite. Then R e ~ if and only if

M e A (a,b) for any pair of analytically independent elements a,b.

Proof: Note that b e @ for any prime P of height 1 or 2, the first case

being obvious and the second using Corollary 4.7. Thus for any I, the primes of

height I or 2 in A (I) are identical to those in A (I). Since A (I) ~ A (I),

we see that the only way to have A * (I) # A-~ (I) is to have M e A*(1) ~(I). Now

if ~(1) = I then for some a e R we have (a) c I ~ (a). As R is normal,

(a) = (a) so that I = (a), and hence M @ A*(I) since R is a Krull domain and

height M > I. That is, A (I) = A (1) whenever ~(1) = i. Now suppose ~(I) = 3.

By Proposition 4.1, M e A (I). Thus A (I)=A (I) whenever ~(I) =3. Since

(I) <__ height M = 3, the only remaining case is ~ (I)= 2. Let J = (a,b) be a
 66

minimal reduction of I. Then a,b are analytically independent. Suppose now

that for any analytically independent pair a,b, we had M e A (a,b). Then

M e A (J)=A (I) so that A (I)=A (I) whenever ~(I) = 2, showing that R e @.

Conversly, suppose R e ~ and consider a,b analytically independent. By Lenmaa

8.5, M c A* ((a2,b 2) +M(a2,b2)) = A-'~((a2,b 2) +M(a2,b2)) =A*(a2, b2). However

(a2,b 2) ~ (a,b) 2 c_ (a2,b 2) so that clearly A*(a2,b 2) = A-* ( ( a , b )2) = A -* (a,b). Thus

M e A (a,b) as desired.

Having investigated when A (I)= A (I) for all i, we now look at two results,

the first assuming more, the second assuming less.

PROPOSITION 8. Ii. Let (R,M) be a 2-dimensional local domain. Then the following

are equivalent.

(i) R is a U.F.D.

(ii) For any ideal l of R, A(I,I) =A(I, 2) ..... A (I)=~(I,i) =A(I,2) .... = A (I).

(iii) For any ideal I, Ass(R/l) = A (I).

Proof: (i) ~ (ii): Any prime minimal over I is in each set in (ii). Thus we

need only worry about M. Claim: If J is an ideal, then M is a prime divisor

of J if and only if J is not principal if and only if ~ is not principal.

Applying this claim to J = I n , n = 1,2,..., shows that if M is in any set in (ii),

then it is in all of them. To prove the claim, the first equivalence is aneasy

exercise using primary decomposition and the fact that R is a U.F.D. For the

second equivalence, we have (using that R is normal) J not principal if and only

if ~(J) > I if and only if ~(~) > i if and only if ~ is not principal. This

proves the claim, and so (i) ~ (ii).

(ii) -# (iii) is in~nediate since A(I,I) is Ass(R/l).

(iii) ~ (i): We first claim that R is normal. If not, by Lemma 8.4, for some

0~ a e R we have M e A*((a)+M(a)). Thus, for large n, M is a prime divisor

of ((a) +M(a)) n. By (iii) and the fact that (a)n ~ ((a)+M(a))n ~ (a) n c_ (an),

we see that M e A ((a)). Proposition 3.19 now shows that M e A (I) for any ideal

140 in R. By (iii), M e Ass(R/l) for any ideal I~0 in R. This last is

 67

impossible since we may let I=P be a height i prime. This contradiction shows

that R is normal. Now if R is not a U.F.D., let Q be a height i prime of

R which is not principal. By Corollary 4.7, M e A (Q). By (iii), M e Ass(R/Q).

This is impossible. Thus R is a U.F.D.

PROPOSITION 8.12. Let R be a locally quasi-unmixed Noetherian ring. Then the

following are equivalent

i) R is Cohen-Macaulay.

ii) A (I) = A (I) for every ideal of the principal class.

Proof: (i) ~ (ii): Let I be in the principal class. The argument used in

[KI, Theorem 125] shows that i can be generated by elements x I , -.., x n with

height(x I,..., xi) =i, i = l,...,n. As R is Cohen-Macaulay, x I,..., x n is clearly

an R-sequence. Using [KI, Exercise 13, page 103] we easily see that A (I)=

Ass(R/l)= [PIP is minimal over I} ~ A*(1) c A*(1). Thus (ii) holds.

(ii) ~ (i) : Suppose that R is not Cohen-Macaulay. Let P be a maximal prime with

height P > gr P = n , and let x I,..., Xn be an R-sequence with P e Ass(R/l) with

I = (xI ,..., Xn). Again using [KI, Exercise 13, page 103], we have P c A (I).

However since ~(Ip) does not exceed the number of generators of ~, we have

_ ~(~).
height P > n > By Proposition 4.1, P ~ A (I). As x I,..., x n being an

R-sequence shows that I is in the principal class, we are done.

 CHAPTER IX: Conforming Relations

Recall that primes P ~ Q in a domain R are said to satisfy going down if

for any integral extension domain T and any prime q of T lying over Q, there

is a prime p of T lying over P with p ~ q. Since going down is well known

to hold for P ~ Q if R is normal, one easily sees that in the above definition

it suffices to consider the case T=R, the integral closure of R.

PROPOSITION 9.1, Let P be prime in a Noetherian domain R. Let Q be a prime

containing P and such that Q is minimal with respect to the property p c Q

fails going down. Then for any ideal I with p c I c Q, we have Q e A (I).

Proof: We find a prime q of R lying over Q such that no prime contained in

q lies over P. Shrink q to a prime ql minimal over IR. Clearly no prime

contained in ql can lie over P. Thus p c ql ~ R fails going down. By the min-

imality of Q, qlnR=Q" Now use Proposition 3.5.

COROLLARY 9.2. Let P be a prime in a Noetherian domain R. If A (p)= {PI,

then P ~ Q satisfies going down for any prime Q containing P.

Proof: Use Proposition 9.1 with I=P.

The converse of Corollary 9.2 fails. Let P be a non-principal height I

prime in a 2-dimensional normal local domain (R,M). Normality implies that P cM

satisfies going down, while Corollary 4.7 shows that M e ~(P).

COROLLARY 9.3. Let P ~ P' be primes in a Noetherian domain which fail going down.

Then there is a prime Q e A*(P) with p c Q ~ p' and P ~ Q fails going d o w n .

Proof: Shrink P' to a prime Q minimal with the property that p c Q fails

going down. By Proposition 9.1 we have Q ¢ A (P).

COROLLARY 9.4. Let P be a prime in a Noetherian domain R, and let W be a

set of primes of R, each of which contains P, and such that p c p' fails

going down for all P' e W. Then N[P' e W} strictly contains P.

 69

Proof: For P' e W, use Corollary 9.3 to find Q e A (P) with Q ~ P' and
-,
p c Q failing going down. Obviously P~Q. Thus N~P' c W] contains n[Q e A (P)]

PCQ]. As A (P) is finite, this last intersection strictly contains P.

Our next result sharpens our appreciation of Corollary 9.4 by applying it to

a special situation.

DEFINITION. Let P ~ Q be primes with height Q/P = i. We say that Q is directly

above P.

COROLLARY 9.5. Let P be prime in a Noetherian domain. Then p c Q satisfies

going down for all but finitely many of the primes Q directly above P°

Proof: Since any infinite collection of primes directly above P intersect at

exactly P, the result follows from Corollary 9.4.

we wish to give a result about going down whose proof has much in common with

an older result about chain conditions. In order to exhibit the similarities be-

tween these two proofs, we will utilize some abstraction.

DEFINITION• Let Q be a prime ideal and let W be an infinite set of prime ideals

each of which properly contains Q. If for any infinite subset W' of W,

n[Q' e W'} = Q then we call (Q,W) a conformin~ pairo

LEMMA 9.6. Let I be an ideal in a Noetherian ring and let W be an infinite set

of primes each of which contains I. Then there is a conforming pair (Q,W) with

I~Q and W~W .

Proof: Enlarge I to an ideal Q maximal with respect to being properly contained

in infinitely many primes in W • It is straightforward to verify that Q is prime.

If W = {Q' e W*IQ c Q'}, it is easy to see that (Q,W) is a conforming pair.

For primes P ~ Q, we will consider a relation which may or may not hold be-

tween P and Q, writing either P *Q or P ~ Q, respectively.

 70

DEFINITION. We will call * a conforming relation if whenever P ~ Q and ~,W)

is a conforming pair, then P *Q if and only if P * Q' for all but finitely many

Q' ~ W.

LEMMA 9.7. If * is a conforming relation, then ~ is a conforming relation.

Proof: Suppose that P~Q and that (Q,W) is a conforming pair. Let W' = [Q' s w I

P*Q']. We must show that W' is finite. If W' is infinite, then (Q,W') is

clearly a conforming pair, and since P *Q' for all Q' e W', we must have P*Q

since * is a conforming relation. This contradiction shows that W' is finite.

Similar reasoning shows that if P ~ Q' for all but finitely many Q' e W, then

P~Q.

LEMMA 9.8. Let R be a Noetherian ring and let * be a conforming relation. Let

P ~ I be ideals with P prime. Let W = [Q e spec R I I ~ Q and P ' Q ] . Then W

has only finitely many minimal members (with respect to inclusion).

Proof: Let W' consist of all the minimal members of W. If W' is infinite,

then Lemma 9.6 shows that there is a conforming pair (Q,W") with I c_ Q and

W" c W'. Since P*Q' for all Q' ~ W", we must have P*Q. Thus Q e W. This

contradicts that Q is properly contained in the members of W", since W" con-

sists of minimal members of W.

We now see that conforming relations are particularly well behaved.

PROPOSITION 9.9. Let P be prime in a Noetherian ring, and let * be a conform-

ing relation with P*P. Then there is a chain of ideals p = I 0 c I 1 c...c In with

the following property: If P ~ Q with Q prime, and if j is the largest sub-

script with I. c Q, then P*Q if and only if j is even.

3 -

Proof: We let I0 = P and inductively construct the chain. Suppose that Im has

been constructed, and assume m is even (the case that m is odd being syn~etric).

Let W = [Q' e spec R I Im ~ Q' and P ~ Q ' ] . By Lermmas 9.7 and 9.8, W has only

finitely many minimal members. We define Im+ I to be the intersection of those

 71

finitely many primes. Since I is (by induction) a finite intersection of primes

m
q satisfying P * q, clearly Im is proper in Im+ 1 • Thus our chain eventually

stops. Suppose now that I is the largest ideal in our chain which is contained
m
in Q, (and still assume m even). If P ~ Q, Then Q e W and so by construction

we would have Im+ I ~ Q, a contradiction. Thus our chain has the stated property.

PROPOSITION 9.10. Let R be a Noetherian ring, and for each P e spec R suppose

there is a chain of primes p = I 0 c I1 c . . . ~ In, n=n(P). Define a relation in

the following way: For P ~ Q let j be the largest subscript with I. c- Q,

and say P *Q if and only if j is even. Then * is a conforming relation.

Proof: Let P ~ Q and let (Q,W) be a conforming pair. Let !. be the largest
J
ideal in the chain associated with P such that I.j c-- Q. By the definition of

conforming pair, at most finitely many members of W can contain Ij+ 1 (lest

Ij+ 1 ~ Q). Thus Ij is the largest ideal of our chain contained in Q' for all

but finitely many Q' e W. The result is now obvious.

Remark: Propositions 9.9 and 9.10 together characterize conforming relations with

the property that P*P for all P e spec R. To accommodate P~P, we notice

that if * is a conforming relation and if U c spec R and if P*Q is defined

as P*Q for P e U and P~Q for P ~ U, then P*Q is a conforming relation.

Thus we have characterized all conforming relations.

We now consider concrete examples.

DEFINITION. The primes P ~Q are said to satisfy catanicity if height Q =

height P + height Q/P. (In [HM] this was called normality.)

LEMMA 9.11. Let (Q,W) be a conforming pair in a Noetherian ring. Then Q c Q'

satisfies catanicity for all but finitely many Q' e w.

Proof: Let height Q = n and let Q be minimal over (x I ,..., Xn). Let

Q = P0' PI' "''' Pm be all of the primes minimal over (x I ,..., Xn). Suppose that

W '= [Q' e W I Q c Q' fails catanicity}o Then for Q' e W', since height Q' >
 72

height Q ' / Q + h e i g h t Q = h e i g h t Q'/Q +n, the principal ideal theorem shows that

height Q'/(x I ,°.., x n) > height Q'/Q. Thus Q' must contain one of PI'"" Pm °

This is true for all Q' e W'. We have P I N ... N Pm ~ N~Q' e W'}. Since

PlN'''N Pm ~ P 0 = Q' the fact that (Q,W) is a conforming pair shows that W'

must be finite.

PROPOSITION 9.12. Catanicity is a conforming relation in a Noetherian ring R.

Proof: Let P ~ Q and let ~,W) be a conforming pair. Letting a prime denote

modulo P, if we let W' = (q'l q e W} then obviously (Q',W') is a conforming

pair in R'. Let W = [q e W I either Q c~ q fails catanicity or Q' c q' fails cata-

nicity}. By Lemma 9.11 applied to both (Q,W) and (Q',W'), we see that W is

finite.

Now let q e W - W. Then we have both height q = height q/Q + height Q and

height q / P = h e i g h t q / Q + h e i g h t Q/P. Of course we also have height Q ~ height Q/P

+ height P, equality holding exactly when P ~ Q satisfies catanicity. Thus

height q = height q/Q +height Q ~ height q/Q + height Q/P +height P = height q/P +

height P. Therefore, examining the one inequality, we see that P ~ Q satisfies

eatanicity if and only if P ~ q satisfies catanicity for any q e W -W.

To prove the result, we note that if P ~ Q satisfies catanicity, then for

all but finitely many q e W, (the exceptions being in W) P ~ q satisfies cata-

nicity. Conversely if P ~ q satisfies catanicity for all but finitely many

q e W, then since W-W is infinite, P ~ q satisfies catanicity for some

q e W-W, and the preceding shows that P ~ Q satisfies catanicity.

We now show that going down is a conforming relation. The next lermma, a

strengthening of Corollary 9.5, is the analogue of Lermna 9.11,

LEMMA 9.13. Let (Q,W) be a conforming pair in a Noetherian domain. Then Q ~ Q'

satisfies going down for all but finitely many Q' e W.

Proof; Let W ' = [Q' e w I Q ~ Q' fails going down}. By Corollary 9.4, N[Q' e W'}

strictly contains Q. By the definition of conforming pair, W' must be finite.

 73

PROPOSITION 9.14. In a Noetherian domain, going down is a conforming relation.

Proof: Let P ~ Q be primes and let ~,W) be a conforming pair. Suppose that

P ~ Q satisfies going down. If Q c Q' also satisfies going down, then obviously

p c Q' satisfies going down, Therefore Lemma 9.13 shows that P ~Q' satisfies

going down for all but finitely many Q' e W.

Now suppose that P ~ Q' satisfies going do~n for all but finitely many

Q' e W. we may ignore the finitely many exceptions and assume P ~ Q' satisfies

going down for all Q' e W. Our task is to show that P ~ Q satisfies going down.

Therefore, if q is a prime of R lying over Q, and if Pl ' "''' Pn are all the

primes of R lying over P, we must show that q contains one of Pl '''°' Pn "

By going up, for each 0' e W, there is a q' prime in R with q' N R = Q ' and

q C q'. Since P ~ Q' satisfies going down, each such q' contains one of

Pl ''''' Pn " Therefore some Pi is contained in infinitely many such q'. Let

U = [q'I Pi ~ q'}' We claim that ~[q' e U} =q. This will show that Pi ~ q' com-

pleting the proof. To prove the claim, let x e f](q' e U}. Since x is integral

over R, we may consider an expression x ~ + r E_I x ~ ' l + , . . + r 0 e q, with r.l e R

and with ~ minimal among such. Since this expression is in q ~ q', and also

x e q', we have r 0 e q' N R = Q ' e W. As U is infinite, we now see that r0 is

contained in infinitely many members of W, Since (Q,W) is a conforming pair,

r 0 e Q ~ q. Thus x(x ~ ' l + r ~ . I x ~ - 2 + . . . + r l ) e q. As q is prime, the minimality

of ~ shows that x e q as desired.

Recall that for R Noetherian, the prime Q is a G-ideal exactly when either

Q is maximal or dim(R/Q) = i and R/Q has only finitely many prime ideals [KI,

Theorem 146]. Recall also that the domain R satisfies going down if p c Q sat-

isfies going down for all primes p c Q.

PROPOSITION 9.15, Let R be a Noetherian domain and suppose that p ~ Q satisfies

going down whenever Q is a G-ideal and height Q/P = i. Then R satisiies going

down.
 74

Proof: Suppose some p c q does not satisfy going down. We may take a counter-

example with q maximal among all such. We claim that q is a G-ideal. Let

W = [q' e spec R I q c q' and height(q'/q)=I}. If q is not a G-ideal, then W is

infinite and (q,W) is a conforming pair. By the maximality of q, for any

q' e W, p c q' has going down. By Proposition 9.14, p c q must have going down,

which is a contradiction. Thus q is a G-ideal.

Fixing q, we now may assume that p has been chosen to make height(q/p)

minimal. If height(q/p) > I then consider the infinite set U = [p' e spec R I

p c p' c q and height (p '/p) = 17. By Corollary 9.5, there is a p' e U with pCp'

satisfying going down. Clearly p' c q cannot satisfy going down. However

height (q/p') < height(q/p) and we have a contradiction, proving that

height(q/p) = I. This now contradicts the hypothesis.

In [MI] it is shown that if p c Q has going down for any Q and any height 1

prime P, then the Noetherian domain R has going down. This can be combined

with our previous ideas. To illustrate this in an easy setting, we will consider a

catenary Noetherian domain.

PROPOSITION 9.16. Let R be a catenary Noetherian domain and let n be less than

the height of any G-ideal of R. Suppose that p c Q has going down whenever Q

is a G-ideal and height P = n. Then R has going down.

Proof: We first inductively reduce to the case that n = I. For this, suppose the

hypothesis holds for n and let p be a prime of height n-l. Furthermore let

Q be a G-ideal containing p. Since R is catenary and height Q > n,

height(Q/P) > i. Thus W = {P e spec R I p c p c Q and height(P/p) =I} is infinite.

By Corollary 9.5, for some P e W, p c p has going down. Since height P =

height p + I = n, by induction, p c Q has going down. Thus p ~ Q clearly has

going down.

We now have p c Q has going down whenever Q is a G-ideal and height P = I.

The technique in the first paragraph of the proof of Proposition 9.15 allows us to

conclude that p c Q has going down for any height I P and any Q. The result

in [MI] now shows that R has going down.

 75

We close with an example of primes p c Plc P2 in a Noetherian domain for

which P ~ Pl does not satisfy going down, while p c P2 does satisfy it.

Thus the chain of ideals p = I 0 c I I c 12 c . . . c In discussed in Proposition 9.9

has n > 2.

EXAMPLE. Let 0 c PI ~ P2 be primes in a Noetherian domain R, and suppose that

in the integral closure R, at least two primes, Q1 and Q2' lie over PI "

Let c e R with c e QI-Q2 and with c in every prime of R which lies over

P2 " Now in R[X], q = (X-c)R[X] is prime. Let p=qNR[X]. If S = R - [0} and

K = R S , then PS = qs = (X-e)K[X], and we easily see that q is the unique prime

of R[X] lying over p.

Now c e QI' so q c (QI' X)R[X]. Contracting to R, we have p c (PI' X)R[X]

We claim going down fails here. This is obvious, since (Q2' X)R[X] lies over

(PI ' X)R[X] and q ~ (Q2 ' X)R[X] since c ~ Q2 ° On the other hand,

P c (P2' X)R[X] does satisfy going down, since every prime of R[X] lying over

(P2 ' X)R[X] has form (Q,X)R[X], with Q a prime of R lying over P2 " By

choice of c, c e Q, so that q c (Q,X)R[X]. Thus we have p c (PI ' X)R[X] c

(P2 ' X)R[X] with P ~ (PI' X)R[X] failing going down, while P ~ (P2 ' X)R[X]

satisfies going down.

 CHAPTER X: Ideal Transforms

In Chapter 6, we called ideals I and J projectively equivalent, I ~ J, if

for some positive integers n and m in = jm . Obviously, in a Noetherian ring,

l~J implies that A (I) = A (J), so that by Proposition 3.17 we see that A (I)

N ~(J), J ~ I. We give an example showing that this inclusion may be proper.

EXAMPLE. Let (R,M) be a 2-dimensional local domain which is quasi-unmixed but not

unmixed. That is, every minimal prime of R has depth 2, but there is a prime

divisor of zero, Q, in R with depth Q = I. (Such an K was constructed in

[F-R]. A more general construction was recently given in [B-R].). Let O # I = aR

be a principal ideal. If J~I, then since QNR= O, we have JR ~ Q. Thus M

is minimal over JR +Q, and by Proposition 1.14, M e A (JR). Therefore M e A (J)

for all J ~ I. On the other hand, Proposition 3.19 shows that M @ A (I) since R

does not have a depth i minimal prime.

In general, we do not understand the differences between A

- * (I) and N A* (J)
J~ I very well, except in the case that I is principal. We explore that case in

this chapter, and see that it has connections to known results concerning ideal

transforms.

DEFINITION. Let R be a Noetherian ring with total quotient ring Q(R). Let I

be a regular ideal of R. The ideal transform of I is T(1) = {y e Q(R) I yl n ~ R

for some n].

LEMMA I0.i. Let I be a regular ideal of the Noetherian ring R. Then

i) If x e ! is a regular element, T(1) c R

-- x

ii) If I = (x I ..., Xn) with each xi regular, then T(1) = R n...N R

X. X
1 n
~ote: such x. always exist.)
i

iii) If R ~ R with R a flat Noetherian extension of R, then T(IR)=T(1)~RR.

In particular, if S is a multiplicatlvely closed set in R, then T(1) S =

T (IS).
 77

iv) If P e spec R and I ~ P, then T (I) contains a unique prime Q lying

over P, and T(1)Q=~ .

v) If x e I is a regular element, and if y is in the total quotient ring

of R and satisfies ly ~ xT(1), then y e xT(1).

vi) I does not consist of zero divisors modulo xT(1) n R for any regular x e I.

Proof: These are fairly simple. We prove only (v) and (vi). For (v), since ly

is finitely generated, for some large n we have In+ly c xR. As y/x e Q(R), and

In+l(y/x) ~ R, we have y/x e T(1). This proves (v), and (vi) follows easily.

In [Nil, Nishimura investigates when T(1) is either an integral extension of

R, or a finite R-module. We reproduce and extend Nishimura's work, first looking

at the integral case. The next lemma simplifies matters. Recall that z(P) is

the minimal depth of a minimal prime in (~) .

DEFINITION. If R is a Noetherian ring with total quotient ring Q OR), then

R[I] = [ y e Q~)I (R :Y)R~_P for all primes P with z(F)=l}.

LEMMA 10.2. R [I] c R.

Proof: Suppose a/b e Q ( R ) - R . Then (R: a/b)R is a proper ideal of R. If P is

a prime divisor of this ideal, then for some c e R we have P = (bR: ac)= (b-~: ac)

since bRNR=b-~. Thus z(P)=l. As (R : a/b) ~ (~: a/b) c p, we have a/b ~ R [I].

PROPOSITION 10.3 Let (R,M) be a local ring with M regular. The following are

equivalent.

i) T (M) c R.

ii) r~) c R [I].

iii) R does not contain a depth i minimal prime.

Proof: (ii) ~ (i): By Lerm~a 10.2.

(iii) ~ (ii): Clearly r~) = [ye Q~)I M~rad(R : y)}. By (iii), z~)#l, and

we see that T(M) c R [I].

 78

i) ~> iii): We will prove the contrapositive. Thus assume that (iii) fails. Since

T~ )=T~ ) = T ~ ) ~ R R , to show (i) fails it will suffice to show that T(M ) !

-7
R . Thus we may assume that R is complete. As (iii) fails, there is a depth i

minimal prime Q of R. Let x e M be regular, so that M is minimal over

(x) +Q. suppose M k ~ (x)+Q. Now for some s e R-Q and n ~ I, sQ n = 0 . By

Lemma 3.11, we may also assume s ~ (xn) by increasing n if necessary. Now

M2nk ~ ((x)+Q)2n ~ (xn) + Q n Thus M2nks ~ (xn). Therefore s/x n e T(M), but

s/x n @ R since s ~ (xn) = x n R N R , and (i) fails.

COROLLARY 10.4. Let I be a regular ideal in a Noetherian ring. The following

are equivalent:

i) T (I) c R.

ii) T(1) c R [I]

iii) z(P) > i for each prime P containing I.

iv) z(P) > I for each P e A (I).

Proof: (i) ~ (iv): If P e A (I), then I c_ p so that T(P) c T(1) c R using

(i) o Therefore T(Pp) c ~ and Proposition 10.3 gives z(P) ~ I.

(iv) ~=> (iii). We treat the eontrapositive. Thus suppose that I ~_ P and z(P) <__ I.

Since I is regular, clearly z(P) #0, and so z0P) = I, Let q be a depth I

minimal prime in (~) . As (Pp) is minimal over ~ +q, by Proposition 3.18

we see that P e A (I). As z(P) # i, (iv) fails. This shows that (iv) =# (iii).

(iii) ~> (ii): Since T(1) = {y e Q(R) I I e rad(R :y)}, this is obvious from the

definition of R [I].

(ii) ~=~ (i): Irmmediate from Lemma 10.2.

COROLLARY 10.5. Let R be a Noetherian ring Then R [I] = UT(1) over all regular

ideals I such that T(1) c R.

Proof: One containment is given irm~ediately by the preceding corollary, If

y C R [I] let I = (R : y) which is a regular ideal since y e Q(R) ° As y e R [I}

 79

if P is a prime containing I, then z(P) > I. Now the preceding corollary tells

us that T(1) c__ R [I]. As y e T(1), we are done.

COROLLARY 10.6. Let (R,M) be a local ring and let n = min{depth ql q is a minimal

prime in R*}. If I is a regular ideal with depth I ~ n-2, then T(1) ~ R,

Proof: By Corollary 10.4 we must show that z(P) > i for every prime P contain-

ing I. Suppose to the contrary that z(P) = i (z(P) = 0 being impossible since I

is regular). Now ~p) has a depth I minimal prime. Also (R/P) obviously

has a minimal prime of depth equal to depth P. Thus by a result proved in the

Appendix, R has a minimal prime of depth equal to depth P + i ~ depth I + I

n-2 + 1 = n-l, a contradiction.

We remark that for a Noetherian domain R, R [I] = N ~ , z(P) =I. Thus R [I]

is in concept similar to the well known extension R(1)= n~, height P = I. It is

interesting that for (R,M) local, these two extensions differ by only a finite

amount (in some sense) as we now show

PROPOSITION i0.7. Let (R,M) be a local ring. The set {P e spec R I z(P) = I

but height P > i} is finite.

Proof: Suppose that P is in the given set. By Proposition 5.6, gr P = I. If a

is an asymptotic sequence from P, then P e A (aR). By Proposition 3.18, there

are primes q* ~ P * or R* with q* minimal, P */q* e ~ (aR*+q*/q* ) and

P * N R = P. By Len~na 3.14, since R*/q* satisfies the Altitude Formula, we have

p*/q* = * ,
height I. However height P ~ height P > I. Thus P must contain some

minimal prime besides q . If J is the intersection of the other minimal primes,

* * * * p*/q*
then J ~ P . Since J ~ q , only finitely many primes can have height = I

and height P > I. As this is true for each of the finitely many minimal q , we

are done.

We now 5egln to consider when T(1) is a finite R-module. Previously, the

following was known((ll)<==> (iii) due to Nishimura, (i) <==> (iii) due to Ratliff

JR4]).
 80

THEOREM. Let (R,M) be a local ring with M regular~ The following are equivalent.

i) R has a depth i prime divisor of zero.

ii) T(M) is an infinite R-module.

iii) There is an integer n such that M e Ass(R/J) for all ideals J ~ M n,

Our contribution (this being work done with Katz) will be to add the following two

statements to this list of equivalences.

iv) M e A (J) for any regular ideal J c P.

v) There is a regular x e M with M e AA (J), J~xR.

LEMMA 10.8. Let I be an ideal of the Noetherian ring R and let x e I be a

regular element. Then T(I) is a finite R-module if and only if T(Ip) is a finite

~-module for each P e Ass(R/xR) with I ~ P.

Proof: One direction is imediate. Thus suppose that T(~) is a finite ~ - m o d u l e

for each P e Ass(R/xR) with I c p. For each such P, there is an n with

xnT(~)n~ ~ x~ (Artin-Rees Lemma). If m is the maximum of those finitely many

n, then xmT(~)NRp ~ xR for all P e Ass(R/xR) (since if I ~ P, T ( ~ ) = ~ )

from which it follows that xmT(1) N R ~ xR. We claim T(1) ~ Rx "m. Since x e I,

T(1) c R . For a e T(1) choose k minimal with xka e R~ If k > m then

-- x
xka e xmT(1) N R c xR which says xk-la c R, a contradiction Therefore k ~ m,

showing that T(1) c Rx -m. Thus T(1) is a finite R-module.

The thrust of the next proposition is that to understand the finiteness of

T(1), it is enough to understand T(Pp), for primes P containing I. In Propo-

sition I0.I0, our key result, we will examine T~p)~

PROPOSITION 10.9. (Nishimura) Let I be an ideal in the Noetherian ring R, and

let x e I be a regular element. The following are equivalent.

i) T(1) is a finite R-module.

ii) T(Pp) is a finite Rp-module for all primes P containing I.

iii) T(Pp) is a finite ~ - m o d u l e for primes P e Ass(R/xR) with I CP.

 81

Proof: (i) -~- (ii): This is trivial since I c p implies TOP) c__T(1).

(ii) ~ (iii) : Trivial.

(iii) ~ (i): We make use of Lemma i0.8. Therefore let P e Ass(R/xR) with I c_ p.

By (iii) we have that TOPp) is a finite R~-module, and our goal is to show that

T(~) is a finite ~ - m o d u l e . Suppose this is false for some such P. We may assume

that P is a minimal counterexample Letting A = T(Pp), we claim that T(IA) is

a finite A-module. In order to see this, we again use Len~a 10.8. Therefore let

Q e Ass (A/xA) with IA c Q, and let Q N Rp = qp (q e spec R) • Suppose that

Q = (xA : y), y e A - x A . Then qpy ~ x A = x T ( P p ) . Since y ~ xA, Lemma 10.1(v) shows

that qp#Pp . Now Lermna 10.1(iv) shows that AQ= (~)qp=Rq, so that T(IAQ) =

T(Iq). As Q e Ass(A/xA) and IA ~ Q, we see that q e Ass(R/xR) and I c q. By

the minimality of P, we have that T(Iq) =T(IAQ) is a finite Rq=AQ-module. By

Lermna 10.8, our claim that T(IA) is a finite A-module is now proved. As (iii) says

that A is a finite ~ - m o d u l e , we deduce that T(IA) is a finite ~-module. Finally,

since ~ c A c Q(~), obviously T(~) _c T(IA). This contradicts our assumption

that T(~) failed to be a finite ~ - m o d u l e .

LEMMA i0.i0. Let I be an ideal in a local ring (R,M). If R is complete in the

M-adic topology, then it is complete in the l-adic topology.

Proof: Let [Xn} be an I-Cauchy sequence. We may assume that xn+ 1 - x n e In .

Therefore if In = (aI , ..., am) then for k >_ i, Xn+ k - x n = (Xn+k - x n + k _ I) +...+

(Xn+ l - x n ) = a l(lrlj ) +...+am(Zrmj ) with j = O,l,...,k-I and rij e I j i = l,...,m.

Being I-Cauchy, [xn} is also M-Cauchy and so Xn ~- x e R in the M-adic topology.

Taking M-adic limits, obviously x - xn = lim (Xn+k - Xn) = a~(limZr..) +...+

a (limZr .). We note that lim Zrij exists in R since rij e I j c M j. Thus
m k+~ mj k+~

clearly x- x e in showing that x +x in the l-adic topology.

n n

PROPOSITION i0.II. Let P be a regular prime in a Noetherian ring R. The follow-

ing are equivalent.

i) T(Pp) is an infinite ~ - m o d u l e .
 82

ii) (Rp)* contains a depth I prime divisor of zero.

iii) There is an m > i such that for any regular ideal J c p(m), p e Ass ~/J).

iv) P e A (J) for any regular ideal J c_ p.

v) There is a regular element x e P with p eNA (J), J ~ x R .

vi) Either height P = i or there is a regular element x e P with

P e A*(xkT(P) NR) for all large k.

Proof: (i) ~ (ii): For this, we may assume that R is local at P. Letting R

be the P-adic completion, we have R a faithfully flat extension of R. Since

T(P) is an infinite R-module, T(P * ) = T (PR*) =T(P) D R * is an infinite R * -module.

, *
Now by Lemma I0.I0, if x e P is a regular element, then R is also xR -adic

complete. By [Ma], T(P )/xT(P ) is a finite R module, Thus by [Mr, Len~na

p. 212] we must have NxnT(p *) ~0.

Therefore we may select 0#y c ( N x n T ( p * ) ) A R *. A well known corollary of the

Artin-Rees Lemma gives us that for some k and large n, (xn : y) , c__ (0 : y) , +
R R
x n - ~ *. As y e x n T(P * ), we see that (xn : y) , is P * primary. Thus (0 : y) , +
R R
xn-~ * is P* primary. Since y # 0, this obviously shows that R* has a depth I

prime divisor of zero. Thus ( i ) ~ (ii).

(ii) ~ (iii): For this we again may localize at P (so that P(n) = pn) and pass

to the completion ~) . Then Lemma 1.13 gives that (ill) holds.

(iii) ~ (iv) ~ (v) are obvious,

(~) ~ (i): Suppose that (v) holds but that T(Pp) is a finite ~-module. We will

derive a contradiction. We have T(Pp) c_ ~ (the integral closure of ~ in its

total quotient ring). This easily implies that T(Pp) = (T(P)nR)p . Now for any

k >_. l, x ~ c_ k ( T ( P ) N R ) A R c xkRNR=x~, so that xk(T(P) N R ) N R ~ x R . By (v),

P e A*(xk(T(P)NR)AR). Localizing, we have Pp e A*(xkT(Pp) n ~ ) . As T(Pp) is

a finite ~ - m o d u l e (our assumption), it is an easy consequence of the Artln-Rees

Lena that for large k and all n >_ i, (xkT(Pp) N ~ ) n = x n k T ( P p ) N ~ • Since

Pp e A*(xkT(Pp) NRp), for large n we have Pp a prime divisor of xnkT(Pp) n R p .

This contradicts Lemma 10.1(vi). Thus ( v ) ~ (i)°

 83

(iv) --~ (vi): Suppose (iv) holds. If x e P is a regular element and if xkT(P)N

R c p, then (iv) implies p ~ A*(xkT(p)f7 R) and we are done. Therefore we need

only worry about the case that x ~ ( P ) N R ~ P. Then xkT(pp) N R p = ~ . Thus

x -k ~ T(Pp) so that for some m, Ppm c- XkRp . This shows that height P = i, prov-

ing (vi).

(vi) ~ (i): If height P = I then clearly (iv) holds, so (i) holds. Suppose that
k*
P e A*(xkT(p)n R) for large k. Then Pp e A (x T(Pp)N Rp) and the argument em-

ployed in proving (v) ~ (i) works again.

k.
Question: Is P e A ~ (x~), x e P regular, k large, equivalent to the conditions

in the preceding proposition?

We may now add another equivalent statement to the list in Proposition 10.9,

analogous to statement (iv) of Corollary 10.4.

COROLLARY 10.12. Let I be a regular ideal in a Noetherian ring. The following

are equivalent.

i) T(1) is a finite R-module.

*
ii) T(Pp) is a finite U - m o d u l e for all. P e A (l).

Proof: (i) ~ (ii) is by Proposition 10.9. Suppose now that (i) fails. Then by

Proposition 10.9 there is some prime P

containing I such that T(Pp) is an in-
*
finite ~-module. By Proposition i0.II, P e A (I), showing that (ii) fails.

LEMMA 10.13. Let P be a regular prime in a Noetherian ring R. Then T(Pp) is

an infinite U - m o d u l e if and only if for some Q e Ass(R) with Q c P, T(P~,) is

an infinite ~,-module, the primes denoting modulo Q.

Proof: Suppose T(Pp) is an infinite ~,-module. By Proposition I0.Ii, (Rp) has

a depth i prime divisor of zero, say Q . Let Q*A~=Qp . Then Qp e Ass( u )

(so Q c p and Q e Ass R). Now ( ~ , ) * = (~/Qp) * = (Rp) * /Qp(Rp) and Q /Qp(~)

is a depth I prime divisor of zero in this ring. Thus T(P~,) is an infinite

~, -module. The converse is similar.

 84

LEMMA 10.14. Let R ~ R be a flat extension of Noetherian rings. Let P be a

regular prime of R and let P e spec R with P n R =P. If

T(P^) is an infinite
P
R^-module, then T(Pp) is an infinite b - m o d u l e . Furthermore, if R is a faith-
P
fully flat extension and if T(Pp) is an infinite Rp-module then for some

e spec R lying over P, T(P^) is an infinite R^-module.

P P

Proof: For the first assertion, we have that P satisfies (iv) of Proposition I0.ii

Now if J c p with J a regular ideal of R, we see that P e A (JR). Thus

P e A (J) and we use Proposition I0.II.

For the second assertion, we let S = R -P and note that (R)s is a faithfully

flat Rs-module. Since T(Ps) is an infinite Rs-module, T(PRs) =T(Ps) ~ S RS is

an infinite Rs-module. By Proposition 10.9, there is a prime PS e spec RS with

PRs c PS such that T(P^) is an infinite R^-module (using ~ S ) ^ =R^). Clearly

-- P P PS P
NR=P and we are done.

PROPOSITION 10.15. Let P be a regular prime in a local ring (R,M) with comple-

tion R . Then T(Pp) is an infinite b

__- m o d u l e if and only if there are primes
* * * * * * * *
Q c p in R with Q e Ass R , P ~ R = P and height P /Q =I.

Proof: Suppose first that such Q c p exist. Let J be a regular ideal of

with J c p. Clearly P is minimal over JR +Q , so by Proposition 1.14,

P * e A* (JR*) . Thus P e A * (J) and so we use Proposition i0.II. Conversely,

suppose that T(Pp) is an infinite b - m o d u l e . By Lemma 10.14, there is a

P e speo R with P NR=P and T(P ,) an infinite R , module. By Lermna 10.13

P P
there is a Q e Ass(R ) with Q c p and ( w i t h p r i m e s d e n o t i n g modulo Q )

T ( Pp~'~ ) an infinite R .' -modu i e. By Proposition I0.II~ P ' e A (J) for every
~ e ~

ideal J~0 of R ' (since R ' is a domain). Let x~0 be any element of P '

Since R*' is analytically unramified, by [SO], forlarge m, J = (xmR*') isnormsl,

(jn=jn all n). As P ' e A (J), for large n, P ' is a prime divisor of jn=jn

= (x~lR*'), Thus P ' e A (xR ' ) . However a s R is complete, R*' satisfies

the altitude formula so that Lemma 3.14 shows that height P ' = i. Thus

height P*/Q* = I as desired.

 85

COROLLARY 10.16. Let P be a regular prime in a local ring (R,M). Then (~)

contains a depth I prime divisor of zero if and only if there exist primes
* , * * , * p*/Q*
Q c p of R with P nR=P, Q e Ass(R ) and height = i.

Proof: Immediate from Proposition 10.15 and I0.II.

we do not know of a direct proof of the preceding. For the analogous result

for minimal primes, we do not need ideal transforms (hence we do not need P regular)

as we now show.

PROPOSITION 10.17. Let P be a non-minimal prime in a local ring (R,M). Then

(Rp) contains a
depth 1 minimal prime if and o n l y i f t h e r e exist primes Q c p
* * * * *
in R with P NR=P, Q minimal, and height P /Q =I.

Proof: Suppose such Q c p exist. Let x e P with x in no minimal prime of

* * * * * --* *
R. Then x ~ Q and so P is minimal over xR +Q , so that P e A ( x R ) by

Corollary 3°i3. Proposition 3,18 gives P e A (xR), and s o Pp e A ( x R p ) . Thus

gr Pp=l, and so P r o p o s i t i o n 5 . 6 shows t h a t z ( P ) = Z(Pp) = 1 as d e s i r e d . Conversely

suppose that z(P) = I . With x as a b o v e , Corollary 5 . 1 2 shows t h a t P e A (xR)

( s i n c e x i n no m i n i m a l p r i m e g i v e s g r xR > 0 ) . By P r o p o s i t i o n 3.I8 used twice, there

are primes Q * c p* of R* with P* N R=P, Q * minimal, and P * /Q * e ~ (xR* + Q * /Q * ).

As R /Q satisfies the altitude f o r m u l a , Lemma 3 . 1 4 g i v e s t h a t h e i g h t P * / Q * = 1.

COROLLARY 10.18. Let P be a regular prime in a local ring (R,M). Then T(Pp) ~_
-- * * * * *
if and only if there are primes Q ~ P in R with P N R = P , Q minimal

and h e i g h t P /Q = 1.

Proof: In~ediate from Propositions 10.17 and 10.3.

PROPOSITION 10.19. Let I be a regular ideal in a local ring (R,M). Let

n = m i n [ d e p th q*lq* is a minimal prime in R }. Let m = m i n ( d e p th Q*IQ* e Ass0R )).

a) If n= I then M e A (I) and T(1) ! ~.

b) If n =2 then either M e A (I) or T(1) ~_ Ro

 86

c) If m = i then M e A (I) and T(1) is an infinite R-module

d) If m= 2 then either M e A (I) or T(1) is an infinite R-module.

Proof: (a): Since n = i, z0M) = I. Proposition 5.6 and Corollary 5.12 show
-4¢
M e A (I). Corollary 10.4 shows T(1) ~ R.

(b): Let q be a depth 2 minimal prime in R . Suppose that M @ ~(I). Then

M @ ~(IR*) by Proposition 3.18, so that Corollary 3.13 shows that M is not

, , , , * , ,
minimal over IR + q . Thus there is a prime p with IR + q ~ p c M . As

depth q * = 2, height p*/q*= i. Letting p = p NR, Corollary 10.18 yields

T(pp) ~_ Rp . By Proposition 10.3, z(p) = i. As I c_ p, Corollary 10.4 gives

(c): Use Propositions I0.Ii and 10.9.

(d) : Let Q * e Ass(R*) have depth 2. If M ~ A*(1) then M* ~ A*(IR*) and

Proposition 1.14 shows M is not minimal over IR + Q . Let P be prime with

* * * * * p*/Q* *
IR + Q c p c M . Depth Q = 2 implies height = i. Letting p=P mR, we

use Propositions 10.15 and 10.9 to see that T(1) is an infinite R-module.

We proceed to prove analogues of Corollaries 10.6 and 10.5.

COROLLARY 10.20. Let (R,M) be a local ring and let n = m i n [ d e p t h Q*IQ* e Ass(R*)}.

If I is a regular ideal of I with depth I ~ n - 2, then T(1) is a finite

R-module.

Proof: If false, then by Propositions 10.9 and 10.15 there are primes Q c P in
* * = *A
R with Q e Ass(R*), height P*/Q* I and I c p R. As R*/Q* is catenary,

depth Q = d e p t h P + i <_ depth IR + l = d e p t h I+I <_ n - l , contradicting the dell-

nition of n.

R [I] was defined in terms of primes with z(P) = i, that is, with ( ~ ) * having

a depth i minimal prime. We now define R <I> analogously.

 87

DEFINITION. Let R be a Noetherian ring with total quotient ring Q(R).

R <I>= {y e QOR) I(R: y) ~ e for all primes P such that ( ~ ) * contains adepth Iprime

divisor of zero}.

PROPOSITION 10.21. Let I be a regular ideal in a Noetherian ring R. Then

T(1) c R (I> if and only if T(1) is a finite R-module. Also R <I>=UT(1) over

regular ideals I such that T(1) is a finite R-module.

Proof: Suppose that T(1) is a finite R-module. By Propositions 10.9 ((i)~(ii))

*
and I0.ii ((ii)~(i)), for any prime P containing I, ( ~ ) does not have a depthl

prime divisor of zero. As T(1) = [y e Q(R) II ~ rad(R: y)}, clearly T(1) ~ R <I>

For the converse, suppose that T(1) is an infinite R-module. Then Proposition 10.9

((ii)~(i)) and I0.II ((i)~(v)) show there is a regular element x e P ~ I with

P e A (xR). Thus for some integer n and c e R, P = (xn: c). Now y = c/x n e Q ( R ) ,

and ly ~ Py ~ R, so that y e T(1). On the other hand, since (R: y ) = P and

since ~ ) has a depth I prime divisor of zero (Proposition i0.II), we have

y @ R <I). Thus T(1) ~ R <I>.

Clearly UT(1), T(1) finite, is contained in R~IIL~. For the reverse inclusion

say z e R <I> and let I = (R: z). As z e Q(R), I is regular. Since z c R <I),

if P is a prime containing I, then ~ ) has no depth I prime divisor of zero.

Propositions I0.9 and i0.II now show that T(1) is a finite R-module. Thus

T(1) c R <I>. As z e T(1), we are done.

Remark: In [B3], Brodmann shows that in a local ring (R,M), if the characteristic

of R/M is either 0 or a regular element of R, and if n =min[depth Q*I

* ,
Q ~ Ass(R )}, then U [T(1) Idepth I ~ n - 2} is a finite R-module. This is

stronger than our Corollary 10.20. We ask a question, which if true, is stronger

still.

Question: If (R,M) is a local ring, is R <I> a finite R-module?

Remark: In [Sc], the following result is reported. Let (R,M) be a local ring.

The following are equivalent: (a) There is a Q c Ass(R ) with depth Q < n,
 88

(b) There is an integer k with M e Ass(R/I) for every ideal I ~ Mk with

height I ~ 0. This is an extension of our Proposition i0.ii ((ii) ~ ( i v ) ) . We do

not know if Proposition i0.II ((ii) <==>(v)) can also be extended. We now present a

major question related to this.

Question: Suppose the sequence Xl,..., Xn in a Noetherian ring R is called a

strong asymptotic sequence if (Xl,... , Xn) •R and f o r i= 1,...,n, x i ~ n A (J),

J~ (x 1 ..., xi_i). Our q u e s t i o n is, if (R,M) is a local ring, will xI ... x

' ' ' n

be a strong asymptotic sequence if and only if height((xl, ..., Xn)R +Q /Q ) = n

for all Q e A s s ( R )? This is true if n= 1 or n = 2. The c a s e n= 1 is trivial

since UA (J), J ~ 0 equals Ass(R). For n=2, let x I, x 2 be a strong asymptotic

sequence in (R,M). Suppose also that Q e Ass(R ) and height((Xl, x2)R +Q*/Q*)

< 2. We w i l l derive a contradiction. Since x1 is regular, this height is 1.

Suppose P is prime in R with (x I , x2)R +Q ~ P and height P /Q = i. If

P=P NR, Corollary 1 0 . 1 6 shows t h a t (Rp) contains a depth 1 prime divisor of

zero, so that Proposition i0.II ((ii)~(iv)) gives P eAA (J), J ~ X l R . Since

x2 e P ~R=P, we have contradicted that xI , x2 is a strong asymptotic sequence.

Conversely, suppose height((xl, x2)R +Q /Q )= 2 for every Q e Ass(R ). Then

x I ~UAss(R ), so that x I ~UAss(R)= nA (J), J ~ 0 . Thus xI is a strong asymp-

totic s e q u e n c e . If x1, x 2 is not a strong asymptotic sequence, then let

P eNA (J), J ~ x l R with x 2 e P. Clearly x I e P, making P regular. By Propo-

sition I0.Ii ((v)~(ii)), (Rp) has a depth i prime divisor of zero. By

Corollary 10.16, there are primes Q c p in R with P NR=P, Q e Ass(R )

and height P*/Q* = i. Since (x I , x2) c p c p* , we get height(x I , x2)R * + Q * /Q * =

height P /Q = 1, a contradiction.
 CHAPTER XI: Miscellaneous

Rees' Valuations

Our first topic is due to Rees [RslI. The crucial primes u s e d i n P r o p o s i t i o n l l . 5

are the ones mentioned in Proposition 3.18(iii), and this material will produce a

characterization of projective equivalence. We offer these facts as our excuse for

reproducing this lovely mathematics.

DEFINITION. Let I be an ideal of an arbitrary ring R, and let x e R. Then

Vi(x) = n < ~ means x e In - In+l , while Vi(x) = ~ means x e N I n , n = 1,2,3, . . . .

PROPOSITION II.I. Let I be an ideal in any ring R and let x e R. Then Vl(X ) =
Vl(Xn)
lim - - exists (possibly being i n f i n i t e ) .
n

VI (xn)
Proof: Let ~ = lira sup - - Let ~ < ~ and ~ < i. We will show that for all
n

Vl(Xn)
sufficiently large n, ~ ~. Since ~ can be arbitrarily close to l, and

can be arbitrarily close to ~, this will show that VI(E ) exists.

Vi(xm)
Since ~ < ~, there is an with - - > ~. Choose ~ an integer with
m
~+i > ~' and consider n > ~m. Suppose that km< n < (k+l)m with k an integer.
k
Note that k ~ ~ so k--$~~ b. Since Vl(Xm ) >__m~, we clearly have Vl(Xn ) >__

Vl(Xn)
Vl(xkm ) ~ k V l ( X m ) ~ km~. This, together with n < (k+l)m, gives n
- - > m (k+l)
k
k+i $ ~ 8~° As this holds for all n ~ ~m, we are done.

PROPOSITION 11.2. Let k >__ 0 be an integer if x e Ik, then V--i(x) >__ k. If

~i(x) > k, then x e Ik.

Proof: If x e i k, then we have xS+l+alxS+...+as+l=0 with a.l e iki . This

V m
I (m-s)k. I (x) (m-s>
easily shows that for m > s + I, xm e Thus > k. Letting
-- m -- m

m+~, we see that VI(X) >.k-

v I (xn)
If V-i(x) > k, then for large n we have n > k so that x n e Ink . Thus
 90

xn + b l x n - l + . , . + b n = 0 where b n = -xn e (Ik) n and the other coefficients are 0.

This shows that x e Ik.

What happens in Proposition 11.2 in case Vl(X ) = k ? We will show that in a

Noetherian ring, it follows that x e Ik, First, however, we show that in a non-

Noetherian ring this need not by the case. For this, consider K(X,Y), with K a

field and X and Y indeterminants. Let w be a valuation on K(X,Y) with values

in Zx Z ordered lexicographically. Let w ( K ) = (0,0), w ( X ) = (I,0) and w ( Y ) = I I , I).

Let W he the valuation ring of w and let I=YW. Since w(X) < w(Y), X 4 I.

Since for all n > I, (n-l,n-l) < (n, 0) < (n,n), we have X n e in-I - in. Thus

VI(Xn) = n - I and so ~I(X) = I. However, I is principal and W is integrally

closed, so ~ = I, and X ~ ~.

LEMMA 11.3. Let R be a Krull domain and let I =uR be a principal ideal. Let

PI''''' Pr be the height I primes of R containing u, and let vi, i = l,...,r,

be the valuation associated with the D.V.R. R Let e i = v i ( u ). Then for x e R,

Pi
v . (x)
V--t(x)=min{e l ~ t i =l,...,r}.
1

Proof: Let ~=min{ e ~ [ i= l,...,r} and let VI(X ) =k. We claim that ~-i < k
I

< ~. Since x e Ik = u ~ , we have v.(x) > vi(uk) = k e i, showing ~ > k. Now let

m be an integer with ~-I < m ~ ~. Then for i = l,...,r, we have vi(um ) = m e i

~e i ~ v i ( x ) . This shows that x e umR=I m so that k = V l ( X ) ~ m > ~-i. Our claim

is now proved. Applying it to xn and noting that v.(x n) = nvi(x), we have

• l

n~-I < Vl(Xn) ~ n ~ . Dividing by n and letting n+~, we have ~l(X) = ~.

LEMMA 11.4. Let I be an ideal in any domain R, and let the domain T be an

integral extension of R. Let J = IT. For x e R, Vl(X) =VJ(x).

Proof: Since In ~ j n clearly V--l(X) ! V--j(x). For the reverse, we first claim
_ jn
that jnNR c In . For this, let y e nR, writing y = alt l + . . . + a m t m with

a. e In and t. e T. Let w I ,..., w k be module generators for T l = R [ t l , . . . , tm]

l i

in
over R. Since ywj e InTl , ywj = bjlW I +...+bjkW k with bj~ e A determinant

argument now proves the claim.

 91

Consider ~ < ~j(x) with ~ rational. There are large integers n with

Vj(x n) ~ n~ and with n~ integral. Thus xn e JnSNR ~ In~. By Proposition 11.2,

~l(X n)
~l(X n) ~ n~. Thus ~i(x)= n ~ ~" Letting ~ + VL(x) , we see that V--I(x)

Vj(x), completing the proof.

PROPOSITION 11.5. Let I be an ideal in a Noetherian domain R. Let ~=R[t'l, lt]

be the Rees ring of R with respect to I. Let u = t-I . Let PI'''" Pr be the

height I primes of ~ which contain u, and for i=l,...,r, let v. be the

l
valuation associated with the D.V.R. . Let ei Then for x e R,
Pi = vi (u).
V--I(X) =min{Vi(x) I i = I .... ,r}.
e.1

Proof: Since um~NR=l TM, for all m >__ 0, clearly Vl(Xn ) =Vu~(X n ) for all n>__0.

Thus for x e R, Vl(X ) =Vu~(X). Now by Lerm~a 11.4 we see that Vl(X ) =Vu~(X ) . We

now use Lemma 11.3.

COROLLARY 11.6. If I is an ideal in a Noetherian domain R and if k > 0 is an

integer, then x e Ik if and only if VI(E ) ~ k.

Proof: According to Proposition 11.2 we must only show that if Vl(X) = k, then

x e Ik. With notation as in Proposition 11.5, we have vi(x ) ~ eik= vi(uk),

i= i, ...,r. Thus x e uk~NR=7.

We now drop the assumption that R be a domain.

PROPOSITION 11.7. Let R be a Noetherian ring with minimal primes ql ' "" "' qs "

Let I be an ideal of R and let x e R. For i=l,...,s let li=l+qi/q i and

let xi=x+qi • Then V-'I(X)=min{~l (xi) I i=l,...,s}.

I

Proof: Let ~ <_ min{Vli(Xi)} with B rational, and let n be a positive integer
-- n
with n~ integral. Now n~ <_ min[Vl.(X i)}. Since R/qi is a domain, Corollary 11.6
1 n~ n
shows that x n e InD'l Thus I n~l reduces (I i ,xi) . By Lermaa 3.6, Inp reduces

(In~,xn), showing that x n e InD . By Proposition 11.2 ~l(X n) >__n~. Thus V--l(x)=
 92

~I(X n)
> ~. This gives ~l(X) > min[~l.(Xi) I i = I,.. sl.
i
Now suppose y < VI(X ) with y rational. For infinitely many positive inte-

gers m, VI(X m) > m y and my is integral. Thus xm e Imy so that for i= l,...,s,

xim e I.
mY,l giving Vl.l(x~) ~ m y . As m+~, we have Vii(xi) ~ y , which shows that

COROLLARY 11.8. Let I be an ideal in a Noetherian ring R. For x e R, and

k ~ 0 an integer, V--l(x)~ k if and only if x e ik"

Proof: This follows easily from Corollary 11.6 and Lemma 3.6.

COROLLARY 11.9. Let I and J be ideals in a Noetherian ring R.

i) l=J if and only if Vl(X ) =Vj(x) for all x e R.

ii) in = jm if and only if m~l(X) = nS(x) for all x e R.

Proof: i) Suppose Vl(X) =VJ(x) for all x e R. Using Corollary 11.8 we have

x e ~ if and only if ~l(x) ~ I if and only if ~j(x) ~ i if and only if x e ~.

Conversly, suppose that I=~. Then for all k ~ I, Ik=l-k=J-k=Jk. Suppose now
-

that for some x, (x) > Vj(x), We will derive a contradiction. For sufficiently

large n, there will be an integer k such that ~l(X n) =n~l(X) ~ k > n~j(x)=

~j(xn). By Corollary 11.8, xn e ik but xn ~ jk, the desired contradiction.

ii) We first claim that for any n ~ 0, and any x e R, nVln(x)

- = ~I(x). For this,

let $ < V--l(X) with ~ rational. There are infinitely many m with m~ integral

and with Vl(xnm ) > nm~. Thus x nm e Inm~ so that V in(X nm) ~__m~. As m-~oo, we

have Tin(X) >_ ~/n. Therefore nVln(x

) - >_ if(x). The reverse inequality is proved

similarly.

Suppose now that m~l(X)=n~j(x) for all x e R, with n and m integers.

Then V--ln(X)=V--l(x)/n=~j(x)/m=~jm(X)- By (i), In=jm~ The converse is proved

similarly.
 93

A Question of Krull

We will use some of our knowledge of asymptotic sequences to construct an ex-

ample concerning an old question. In 1937, Krull [Kr, p. 755] asked if R ~ T is

an integral extension of domains with R normal and if P ~Q are adjacent primes

in T (that is, height Q / P = I) must PAR c QNR be adjacent in R? In 1973,

Kaplansky gave a negative answer [K2]. Kaplansky's example has R non-Noetherian,

and he pointed out that the construction of a Noetherian example would probably re-

quire a counterexample to the chain conjecture. In 1980 such a counterexample was

found (see [H2]) and shortly thereafter Ratliff used it to construct a Noetherian

counte~example to Krull's question. Although the construction does not require the

use of asymptotic sequences, using them gives a short path to the example.

DEFINITION. The domain R is a G.B.-ring (going between) if any pair of adjacent

primes in any integral extension domain of R, contracts to a pair of adjacent

primes in R.

Two lemmas will proceed our example. The first shifts attention from integral

extensions of R to the polynomial ring R[Y].

LEMMA II.i0° Let R be a Noetherian domain and let Y be indeterminate. Let

p c Q be primes of R with height Q/P > I. In RIY], let p ~ q be primes with

pAR =P and qNR =Q and with height q / p = i. If p contains a monic polynomial,

then R is not a G.B. ring.

Proof: Let p contain the monic polynomial f(Y). In some splitting field, let

f(Y) = (Y-Ul) .°. (Y-Un) , and let S = R [ u I ,..., Un]. Thus R ~ S is an integral ex-

tension. In S[Y], let p' be a prime lying over p. Since f(Y) c p', for some

i we have Y - u i e p'. Thus (Y-ui)S[y ] NR[Y~ ~ po Now (Y-ui)S[y ] NR[Y] is

easily seen to be the kernel of the map R[Y] +R[ui]. Since this kernel is contained

in p, the images under this map of p c q are two adjacent primes of R[ui],

whose contractions to R, P c Q, are not adjacent.

 94

LEMMA ii.II. Let ~,M) be a local domain of altitude at least 3. Let a,b be

elements of R with height (a,b) = 2. Suppose that R[a/b] contains a height i

prime M' lying over M. Then with X an indeterminate, R[X] is not a G.B.-ring.

Proof: Let x = a/b and pick y e M' with y not in any other prime of R[x]

which lies over M' NR[x]. Then N = M ' NR[x,y] is a height 1 prime of R[x,y].

Let p be the kernel of R[X][Y] +R[x,y] sending X to x and Y to y. Let

q be the inverse image of N. Thus p c q and height q/p = I. Letting

p = p N R[X] and Q = q n R[X], to invoke Lemma ii. I0, we must show that heightQ/P > i,

since y integral over R[x] shows that p contains a polynomial monic in Y.

Now P is the kernel of the map R[X] +R[x] and this map clearly sends Q to

NNR[x]. Thus we have height Q / P = height N N R [ x ] . As M ~ N, MR[x] ~ N O R [ x ] .

By [D, Corollary 2], MR[x] is a prime with height equal to dim R - I ~ 2. Thus

height Q/P ~ 2 and by Lelmna ii. I0 we are done.

EXAMPLE. Let (R,M) be a local 3-dimensional normal domain not satisfying the

altitude formula. Thus R is not quasi-unmixed, so z~) < 3. As R is normal,

z~) > I (Proposition 3.19). Therefore z(M) = 2. Using Proposition 5.6, let a,b

be a maximal asymptotic sequence in R. Obviously height (a,b) = 2. Since

M e A (a,b), by Proposition 3.20, we may assume that R[a/b] contains a height I

prime lying over M. By Len~aa II.II, R[X] is a normal Noetherian domain which is

not a G.B.-ring.

Remark: R[X I ,..., X n] not being a G.B.-ring is thoroughly studied in [R5] and

[B4].

z(P) and Conformin~ Pairs

PROPOSITION 11.12. Let (P,W) be a conforming pair in a Noetherian ring R, and

suppose that z(Q) ~ n + l for all Q e W Then z(P) ~ n .

Proof~ Suppose that z(P) > n - i. Since z(P) = gr Pp , let a I,..., a n be ele-

ments of P whose images in ~ form an asymptotic sequence. We will show that

it is a maximal asymptotic sequence, so that z ~ ) = n. We first claim that for

 95

all but finitely many Q e W, the following is True: If p e UA ((a I,..., ai)),

i =0,1,...,n, and p e Q, then p ~ P. This is easily seen since UA ((a I,..., ai))

i=0,1,...,n, is a finite set, and (since (P,W) is a conforming pair) any p ~P

is contained in at most finitely many Q e W.

By deleting the finitely many exceptions in the above claim, we add the assump-

tion that for all Q e W, p e U A ((a I ,..., ai)) , i = 0, I , . . , n , if p ~Q, then

p _c p. We now claim that for all Q e W, the images of a I,..., an are an asymp-

totic sequence in RQ. If not, then for some Q e W and i=O,l,...,n-I we have

a prime pQ e A ((a I ,..., ai)RQ) with the image of ai+ I in pQ . Note that

ai+ I e p e A ((a I ,..., ai)) and p ~ Q. By the above assumption, p ~ P. Thus

pp e A ((a I ,..., ai)Rp) and the image of ai+ I is in pp . This contradicts that

the images of a I , .. ., an form an asymptotic sequence in ~, and proves our

second claim.

Finally, we claim that P e A ((a I ,..., an)). If not, pick an+ 1 e P with

an+ 1 not in any prime p satisfying p e A (a I ,..., an) and p c p, Our assump-

tion on W easily shows that the images of a I ,..., an+ 1 form an asymptotic

sequence in ~ for all Q e W. Since gr Q Q = z(Q) ~ n + l , this must be a maximal

asymptotic sequence, showing that QQ e A ((a I,..., an+l)RQ) , and Q e A (aI,...,

an+l) , for all Q e W. As W is infinite, this is impossible, proving our final

claim. We now have P e A (a I,..,, an) so that the images of a I ,..., an are a

maximal asymptotic sequence in ~ . Thus z(P) = gr (Pp) = n.

The above argument, due to Heitmann, similarly shows that if grade QQ ~ n + I

for all Q e w, then grade Pp ~ n.

COROLLARY 11.13. Let P be a prime in a Noetherian ring R such that ~ is

quasi-unmixed. Let W = {Q e spec R I P c Q, height Q / P = I}. Then for all but

finitely many Q e W, RQ is quasi-unmixed.

Proof: If W is finite there is no problem. Suppose W is infinite. Then

(P,W) is a conforming pair. Let height P = n. As Rp is quasi-unmixed, z(P) = n.

Let W ' = [Q e W i z(Q) ~ n}. If w' is infinite, then (P,W') is a conforming

 96

pair, and Proposition II.12 shows that z(P) ~ n-l, a contradiction. Thus W' is

finite. Now let W ''= [Q e W 1 height Q > n + l ] . By Lemma 9.11, W" is finite.

Since we always have z(Q) ~ height Q, for any Q e w - (W' UW") we have n+l

z(Q) ~ h e i g h t Q ! n + i. Thus z(Q) = height Q, and RQ is quasi-unmixed.

The recently constructed counterexample to the chain conjecture gives a normal

Noetherian domain with primes p c Q such that height P = 2, height Q/P = i and

little height Q = 2. As another corollary of the above, we show that for a fixed

height 2 prime P in a normal Noetherian domain, there can be at most finitely

many such Q. (If R is not normal, there can be infinitely many such Q.)

COROLLARY 11.14. Let P # 0 be a prime in the Noetherian domain R. Let

W = [Q e spec R 1 P c Q, height Q / P = i, and little height Q = 2}. If W is infinite,

the ~ contains a height I prime lying over P. If R is normal, height P = i.

Proof: Since z(Q) ~ little height Q (see the Appendix), we have z(Q) ~ 2 for

all Q e W. If W is infinite, then clearly (P,W) is a conforming pair. By

Proposition 11.12, z(P) = i (since P # 0 implies z(P) # 0). Thus (~) contains a

depth I minimal prime. By Proposition 3.19, R contains a height I prime lying

over P. The last sentence of the Corollary is obvious.

Question: Suppose for primes P ~ Q in a Noetherian ring, we define the relation

z-catenicity to mean z(Q) = z (P) + z(Q/P). Is z-catenicity a conforming relation?

Prenormality and U A (J), J ~ l

Recalling that an ideal I is normal if I n= I n for all n >_ I, we make the

following definition.

DEFINITION. The ideal I is prenormal if some power of I is normal.

In K[X3,X4,X5], K a field, I = (X3,X4) is prenormal but not normal since

X 5 e ~ - I, while I n = In for all n ~ 2.

PROPOSITION 11.15. Let I be a regular ideal in a Noetherlan ring. The following

are equivalent.
 97

i) I is prenormal.

ii) In = In for infinitely many n.

iii) In = In for all large n.

iv) ~I= -I~ for all large n.

v) for all n.

Proof: (i) ~ (ii), and (iii) =~ (i) are obvious. For (ii) ~ (iii) we use Lermna 8.1

and have (In+~ : I~) = In for all large n and any ~. Since Inl ~ ~ In+9, and

s i n c e by (ii) we may choose ~ such that In+~ = In+~, we see that In c (In+~ : I~) =

In , proving (iii). For (iii) ~ (v), for any n, select m large enough that

In m = Irim. Now (In)TM ~ (In) TM ~ In m = Irim, so that (In)TM = (In)m. By Proposition

8.2(ivy, In ~ In . Since the other inclusion is automatic, we have (v). Obviously

(v)~ (iv), and (iv) ~ (iii) follows from Proposition 8.2(ii).

PROPOSITION 11.16. Let I be a regular ideal in a Noetherian ring R. Then

Ass(In/I n ) stabilizes for large n°

Proof: For large n, (In+l : I) = I n . Suppose that P e Ass(In/I n ) and write

P = (In : c) with c e In . Clearly p c (In+l : cl), and if x e (In+l : cl) then

xe e (in+l : I) = In , so that x e (In: c) =P. This shows P = (In+l : cl). As

cl ~ Inl ~ I n+l , we have P ~ Ass(In+l/In+l). Therefore, for large n, Ass(In/I n )

is an increasing sequence. Also, for large n, Ass(In/I n ) ~ Ass(R/I n ) ~ A*(1).

Since A (I) is finite, the result follows.

DEFINITION. For I a regular ideal in a Noetherian ring, let C (I) denote the

limit of the sequence Ass(In/In), n = 1,2,...

We now see that prenormality is a local condition.

PROPOSITION 11.17. Let I be a regular ideal in a Noetherian ring. Then I is

prenormal if and only if is prenormal for all P e C (I).

Proof: One direction is trivial. Thus suppose C (I) = [PI' "''' Pm ] and that

is large enough that Ass(l-k/lk) =C*(1) and r~i = --~ for all i and
 98

__ m

for all k > n. We claim that I k = Ik for all k > n. If not, pick x e ik - ik.

Then (Ik : x) is a proper ideal. By Lemma 1.2, for some d c R, (Ik : xd) is a

prime P. As xd e I k, P e Ass(Ik/l k) =C~(1). Since x e ~= , there is an

s ~ R-P with sx e Ik, contradicting that (Ik : x) ~ P.

*
Since C (I) is finite, the next result shows that {Q c spec R I IQ is pre-

normal} is open in the spec topology.

PROPOSITION 11.18. Let I be a regular ideal in a Noetherian ring R. Then for

Q e spec R, IQ is not prenormal if and only if there is a P ¢ C (I) with p c_ Q.

Proof: IQ is prenormal if and only if IQ=~ for all big n if and only if

Ass(IQ/IQ) = ~ for all big n if and only if C (IQ) = ~. If C (IQ) # ~, then

there is a PQ e c (~). We easily see that P e C (I) and p c Q. Conversly,

if C (IQ) = ~ then no such P can exist.

PROPOSITION 11.19. Let I be a regular ideal in a Noetherian ring. Then A (I) =

A (I)UC (I).

Proof: One containment is obvious. Thus let P ¢ A (I) - A (I). Therefore

Pp e A * ( ~ ) - ~ ( ~ ) , so that for big n we have P p = (Ip : On) with cn e ~ • If

cn ~ g, then Pp e ~ ( ~ ) contradicting that P 4 ~(I). Thus cn e ~ , and so

Pp ¢ C (~), showing that P ~ C (I) as desired.

In Chapter I0 we looked at NA*(J), J~l. Now we glance at U A (J), J ~ l ,

and see that in many circumstances this union is big.

LEMMA 11.20. Let I be a regular ideal in a Noetherian ring R. If P is a prime

containing I, and i f Ip is not prenormal, then P c UA ( J ) , J ~ I .

Proof: We have ~ for all big n. By Proposition 8.2(ii), ~=~ for all

big n. Thus for big n, ~ ~ ~, and Lermna 8.3 gives Pp e A * ( ~ + P p I p ) . Thus

P e A*(In+pln). As In+pin~l, we are done.

 99

PROPOSITION 11.21. Let I= (aI ,..., an) he a regular ideal in a Noetherian ring

with n > I. Suppose that P is a prime containing I, and that in ~,

a I,..., an are analytically independent. Then P eUA (J), J ~ l .

Proof: Let 2 ,..., an).

I= (aI 2 In ~, a~,..., an2 are also analytically independent.

Now I ~ 12 ~ Y. For any k ~ I, ~ is minimally generated by all monomials of

degree k in a~ ..... a~, while ~k is minimally generated by all monomials of

degree 2k in al,... , an . Thus v(~)#v(~k), so t~at ~# (~)k As ~

( ~ ) k ~ (~), we have ~ (~), for all k ~ I, showing that ~ is not pre-

normal. By Lermna 11.20, P e U A (J), J ~ . However, I ~ 12 ~ I shows that

I~I, so we are done.

COROLLARY 11.22. Let I be a regular ideal in the principal class of a Noetherian

ring. If height I = n > I, then for any prime P containing I, P c U A (J), J ~ I.

Proof: Let I = (al,... , an). If P contains I, then height ~ = n shows that

aI ,..., an are analytically independent in ~, and we use Proposition 11.21.

OOROLLARY 11.23. Let (R,M) be a local domain with R/M infinite. Let I be a

regular ideal with ~(I) = n > i. Then M eUA (J), J ~ l .

Proof: Let - (aI,..., an) be a minimal reduction of I. As aI,..., an are

* ^
analytically independent, Proposition 11.21 gives M eUA (J), J ~ l . However I~I.

Remark: We do not know if this works for a prime other than the maximal ideal of a

local ring, the problem being that ~ a reduction of ~, does not imply that

reduces I. The next section is related.

Loca! 'projective Equivalence

Throughout this section, R will always be Noetherian, and all ideals will be

assumed to have height greater than zero.

LEMMA 11.24. Suppose l~J, with In= Jm , n and m positive integers. Then if

k and ~ are positive integers, Ik = J~ if and only if n/m= k/~.

 1 O0

Proof: Assume that I k = j ~. By Corollary 11.9, (n/m)~j(x) =V--l(X) = (k/~)%(x) for

all x e R. In order to conclude that n/m = k/~, we need only exhibit an x with

0 < Vj(x) < ~. As height J > 0, pick x e J but in no minimal prime. By

Corollary 11.8, V--j(x)~ I > 0. By Lermna 3.11, x ~ O Ij, j= 1,2,3,..., so that

Corollary 11.8 also gives Vj(x) < ~. (Note: finding this x is the purpose of

our restriction that height J be at least i. In fact such an x exists if for

some maximal M ~ J, JM is not in the nilradical of RM . However, this last con-

dition does not localize.)

DEFINITION. If I ~J, let ~ (l,J) = n/m with in = jm.

LEMMA 11.25. Let l~J and let P be a prime containing I. Then ~ ~ Jp and

~ ( ~ , Jp) =y(I,J).

Proof: If in = jm, then m

~ = Jp •

EXAMPLE. Let M and N be distinct maximal ideals. Let I=M2N and J = M N 2.

Then ~=~=J~ and ~ = N 2N = J N " Thus ~~JM and ~~JN" Yet l~J since

if l~J, then Lemma 11.25 would give y ( ~ , JM) =y(l,J) = ¥ ( ~ , JN ). However, ob-

viously ¥ ( ~ , JM) = I / 2 # 2 = y ( ~ , JN).

The preceding example illustrates the main problem in going from local to glo-

bal projective equivalence, as we now see.

THEOREM 11.26. The following are equivalent

(i) i ~J

(ii) ~JM for all maximal ideals M containing in J, and { y ( ~ , JM) I

IN J c M maximal] has size i.

(iii) ~Jp for all P e A (I)UA (J), and [y(IQ, JQ) IQ is minimal over I}

has size I.

Proof: (i) ~ (ii);~> (iii). Both follow from Lermna 11.25. Suppose (iii) holds,

with [y (IQ , JQ) IQ is minimal over I] = In/m}. We will show that in = jm, proving

(i). If false, suppose x e In - J m (the other possibility being similar). Then

 101

(jm x) is a proper ideal, which by Lemma 1.2 can be expanded to a prime

p= (jm xy) for some y e R. Now P e A(J,m) ~ ( J ) and so by (iii), ~~Jp •

This shows that rad ~ = rad Jp , so there is a prime Q ~ P with Q minimal over

both I and J. We already have y(IQ , JQ) =n/m, and so by Lemma 11.25,

y (~, Jp) = n/m• Now Lermna 11.24 gives m .

= Jp As x e In , for some s c R-P,

sx e j m Thus s e (jm x) c ( j m xy) = P, a contradiction,

~(l~n) = B(l,n)

In Chapter I we considered the two sequences A(l,n) = A s s ~ / l n) and B(l,n) =

Ass(In-i/xn), n=1,2,3,... . In Chapter 3, we looked at A(l,n)=Ass01/In),

n = l , 2 , ° .... Finally, we consider B(l,n) =Ass(In-l/In), n = 1,2, . . . . We show that

it is identical to ~(l,n). The key is a lemma due to Katz, which will also give an

easy proof that ~(l,n) is increasing.

LEMMA 11.27. Let I be an ideal in a Noetherian ring R. Then for all n > m > i

(In : I TM)= (In : Im) = in-no

Proof: Since In-m ~ (In : I TM) ~ (l--n: Im), it is enough to show (I-~: I TM) ~ In'm.

First we assume that R is a domain, and let V be any valuation overring of R.

If x e (2n : I TM) then xlmv c Inv = Inv. Since IV is principal, x e In'mv. As

= i n-m
this holds for all such V, and as In'm (N In-mv) N R, we have x e . Now

for an arhitrary Noetherian ring R, we let q be any minimal prime ideal, and let

primes denote modulo q. Then (In : Im) ' ~ (l'n : I 'm ) ~ l'n-mo As this holds for

any minimal prime q, Lermma 3°6 shows that (l--n: I TM) ~ I ~'~m.

00ROLLARY II.28. For an ideal I in a Noetherian ring, ~(l,n) = B(l,n), n = 1,2,

and this sequence is increasing.

I

Proof: Clearly B(l,n) c_ ~(l,n) o Let P e ~(l,n) and write P = (In : c), e e R.

Since ~ P, ~c ~ I n showing that c ¢ (In : 5) = In-l. Thus P e B(l,n), and

~(l,n) =B(l,n) • Furthermore, clearly p c (In+l : c~) while if x e (In+l : c~) then

xc e (ln+l : ~) = in so that x e (In : c) =P. This shows that P = (In+l : ¢~), so

that P e A(l,n+l), showing our sequence is increasing.

 102

We wish to give an example to show that in Proposition I0oli ((ii) ¢=> (iv))

and in Proposition 3.19 ((i) <==> (ii)) it is not enough to only consider prime ideals.

Thus we construct an n-dimensional local domain (R,M) such that M e A (P) for

all primes P ~ 0 (in fact M e A (I) for any ideal I containing a nonzero prime)

and M e Ass(R/P TM) for all m > 2, but such that R has no depth i prime di-

visor of zero. First, an easy lemma.

LEMMA 11.29. Let I be an ideal in a Noetherian ring R and let x e I be a reg-

ular element. Suppose that RN R is a finite R-module. Then T(I) c R [I] if and
x
only if T(I) c R (I>.

Proof: Since R <I> ~ R [I], one direction is obvious. Thus suppose that T(1) c R [I]

Since R [I] ~ R and T(I) ~ R , we have T(1) c R n R x, so that T(1) is a

finite R-module and we use Proposition 10,16.

EXAMPLE. Let n >__ 2. By [HI ] there is a normal Noetherian domain T satisfying

the Altitude Formula with exactly two maximal ideals N1 and N 2 , both of height

n, such that 0 is the only prime contained in N1 n N2 . By a standard gluing

process (See [DL]) there is a local domain ~,M) satisfying the Altitude Formula

with R=T, MT c_R (so that [ = T is a finite R-module), M = N 1 N N 2, and for any

prime P #M of R, exactly one prime of T lies over P.

For P e spec R - [ 0 , M } let Q be the unique prime of T lying over P. Since

Q ~_ N 1 n N 2 , without loss we may assume that Q _c N 1 and Q ~_ N 2 . We claim that N2

is minimal over PT. Suppose that PT c q c N 2 with q prime, and let p = q N R.

Clearly p c_ p, and since Q N R=P, by going up we can find a prime q' of T

with Q c__ q' and q' N R = p . Now q' ~ N 2 since Q ~ N2 . Since q c_N 2, q4q'

and these are two distinct primes lying over p. By the nature of R c T, we must

have p=M, so that q' = N 2 . This shows that N2 is minimal over PT as desired.

Therefore, by Proposition 3.5, M e A (P). (Note: The same argument works for any

ideal I containing P.)

 103

We now wish to show that M ~ Ass (R/PTM) for m > 2. For this, choose

bl e N I - (N21UN2) and b2 e N 2 - (N22UNI) so that b=blb2 c ~ i NN2) -(N21UN22 ).

Since we already have N2 minimal over PT, for any m >_. 2 we have an se T - N 2

and a k >__ I with sNk c_pmT. Using that b ~ NI~N2=M and MT ~ R , weseethat

bsM k c bsN2k _ bpmT ~ p m We claim that bs ~ pm (note bs e R since b e M). If

bs e pm then 6s e ~2' and since s e T-N 2 , we get b e ~2' contradicting

that b ~ N2 • Thus bsM k c pm but bs e R .pm, showing that M e Ass~/pm),

m>2.

Finally we want that R does not contain a depth I prime divisor of zero.

If it did, then by Proposition i0.ii and 10.21 we would have T(M) ~ R (I>. As

is a finite R-module, Lemma 11.29 shows that T(M) ~ R [I]. Thus byPropositionl0.3,

R contains a depth i minimal prime. As R satisfies the altitude formula,

height M = i. This contradicts that height M = n > 2.

Stron$ Asymptotic Sequences

The last paragraph of Chapter X discusses the possibility of developing a con-

cept of strong asymptotic sequences which would stand in relation to prime divisors

of zero, as asymptotic sequences stand to minimal primes. This section will discuss

such a concept. As it was developed too late for major treatment in the text, this

postscript will point out the path, a full exposition appearing elsewhere. I do not

know if the definition of strong asymptotic sequence given here coincides with that

given in Chapter X (it does for sequences of length I or 2).

Notation: Let I ~ J be ideals in a Noetherian ring. (I : J) ~ (I : j2)

(I : j2) c... eventually stabilizes. We denote that stable ideal as I : <J>.

(Thanks to P. Schenzel for the nice notation, and for illustrating the significance

of this ideal.)

LEMMA ii.30. Let I be an ideal in a local ring (R,M). The following are equiv-

alent.

i) There is a Q e Ass R with M minimal over I+Q.

ii) A ITM : <M> ~ 0.

m=l
 104

Further, if R is complete these are equivalent to

iii) There is a k > 0 such that for all m > 0, Im : <M> ~ M k.

Proof: i) ~ i i i ) . If Rad I = M , then Im: < M > = R for all m > 0 and it is

trivial. Thus assume Rad I # M . Then ITM c] ITM : <M>#R, and so by (i), M is

minimal over Im : < M > + Q . Now let ql N...N q n = 0 be a prSmary decomposition of

0 with Rad ql =Q" Select 0 # x e (q2 N...N qn ) - ql " Let ITM : <M) ~ P ¢M, P

prime. Then Q ~ P and so we may pick y c ql - P" As y x = 0, x is in every P-

primary ideal. Since M is clearly not a prime divisor of ITM : <M>, primary de-

composition gives x e Im: (M>. Thus (ii) holds.

In case R is complete, the equivalence of (iii) is easy using [N, 30.1] and

the Krull intersection theorem.

PROPOSITION 11.31. Let I c p be ideals, P prime, in a Noetherian ring R. The

following are equivalent

i) There is a q e Ass(~) with (Pp) minimal over l(Rp) +q .

ii) There is an integer k > 0 such that P e Ass R/J for every ideal J

satisfying Rad J = R a d I and j ~--p(k)

iii) There is an integer k > 0 such that for all m > O, Im : <p> ~_ p(k)

Proof: (i)<==> (iii). Note that ITM : <P> ~ P (k), ~ k

: <Pp> ~ Pp, and

im%)* : <(Pp) * > ~ (Pp) *k are all equivalent. Thus the equivalence of (i) and (iii)

follows from Lemma 11.30.

q~
(i) ~ (ii). Choose k to equal the n of Lermma 1.13 applied to ~ (Pp) . If

Rad J = R a d I then by (i), (Pp) is minimal over J(Rp) +q . If also J ~ p(k),

* *k * *
then J(~) ~ (Pp) . Thus (Pp) is a prime divisor of J(Rp) , and so

P e Ass R/J.

(ii) ~ (iii). If P is not minimal over I, then Rad I =Rad(Im : <P)) and since

P ~ Ass R/(Im : <P>), (iii) follows easily from (ii). If P is minimal over I,

then (iii) holds with k = I, using primary decomposition.

 105

DEFINITION. For I an ideal in a Noetherian ring R, let A.(1) = {P e Spec R[

I c p and there is a k > 0 such that P e Ass R/J for all ideals J satisfying

Rad J = R a d I and j ~P(k))"

Remarks. (a) A.(0) = Ass R.

~) A.(1) c_ A (I) (and so is finite).

(c) If S is a multiplicatively closed set in R and pN S = ~ then

P e A.(1) if and only if PS e A.(Is)° (Since iTM : (p> ~ p(k)

m (k))
if and only if I S : <Ps) ~ (Ps) .

(d) P e A.(1) if and only if there is a Q e Ass R with Q ~ P and

P/Q e A.(I+Q/Q) (if P e A.(1), then there is a q e Ass(~)

with (Pp) minimal over I(~) +q . Take Q e Ass R such that

q N R ? =Qp)o

LEMMA 11.32. Let R c T be a faithfully flat extension of Noetherian rings. Let

I be an ideal of R. If P' e A.(IT), then P' N R e A.(1).

Proof: Let k be as in the definition of A.(IT) applied to P' If Rad J =

Rad I and J ~ (P'N R) (k), then Rad JT = Rad IT and JT ~ p,(k) so that

P' e Ass T/JT, Thus p' N R e Ass R/J,

The converse of Lermna 11.32 is true, but requires the following clever result.

PROPOSITION 11.33. (Schenzel) Let I c J be ideals in a Noetherian ring R.

Suppose for all P e A (J), and for all q e Ass(~) , we have

height I ( ~ ) * + q * /q <__height J ( ~ ) * + q * / q * . Then for all k > 0 there is an m> 0

with Im : <J) ~_ jk.

Proof: Suppose for some k > 0 we have ITM : <j> ~_ jk for all m > 0. Increasing

k does no harm, and so we easily find a P e A * (J) with ~ : <jp > ~_ Jp

k . There-

fore Im(~) * : <J~)*> ! J k -(Rp)

- * . We may assume that P is minimal in A * (J)

with the property that this non-containment is true for some k and all m. We

claim for some q* e Ass (Rp)* , height l ( ~ ) ~'+ q ~ /' q ~~ =height J(Rp) * +q*/q*. To ease
 106

notation, we will assume R = %)*. That is, we will assume 0R,M) is a complete

local ring, that M e A * (J), that ITM : <J> ~ jk for all m > O, and that if

p e A (J) - [M} then for all k there is an m with I~ : <Jp> c_ jkp . Now let
= iTM
Em : <J> and note that E 1 r~ E 2 ~ E 3 ~ .... Fix a k large enough that

Ass R/J k has stabilized to A*(J). Since A*(J) - {M) is finite, our assumption

on R shows that for sufficiently large m, the module E + jk/jk localized at

m
p e A (J)- {M} is O. Thus for large m, the only possible associated prime of

E +jk/jk is M. The annihilator of any element of our module therefore has radical
m
equal to M. Since our module is finitely generated, its annihilator contains a

power of M. Thus for large m, E m + J k / J k has finite length, and as our modules

decrease as m increases, we find that for a fixed large k the ideals E +jk
m
stabilize for large m. Now R is complete in the M-adic topology, hence also in

the J-adic topology. The argument used to prove IN, 30.1] can be used to show that
co
if A E = 0 then for all k > 0 there is an m with E c j k However we are
m m --
m=l
assuming this containment fails for some k. Thus n E # 0. Let x @ 0 be in that
m
m=l
intersection. A well known corollary to the Artin-Rees Lemma shows that for large

and m > 9, Im :x c ( O : x ) + I m'~, The choice of x shows that for each m

there is an n with jn < ITM : x c (0 : x) + Im-~. We may expand (0 : x) to a prime

q e Ass R. Since I c_ j and jn c_ q* + Im-~ clearly Rad I + q * / q * = Rad J + q*/q*.

*/ * *
Thus height I + q * / q * = h e i g h t J+q q . As M ~ A (J), our initial hypothesis is

contradicted.

LEMMA 11.34. Let I c J be ideals in a Noetherian ring R. Then there is a

k > 0 such that for all m > O, Im : (J) ~_ jk if and only if there is a

P c A,(I) with J c_ p.

Proof: Suppose such P exists. By Proposition ii.31, there is a k > 0 such that

for all m > O, Im : <e> ~ pOk) . As ITM : (p> c_ Im : <J>, and jk c_ p(k), we have
im : <J) ~ j k Conversely, suppose ITM : <J> ~_ jk for all m > O. By

Proposition 11.33, there is a prime p containing I such that for some

q * e Ass(Rp) * , we have height l(Rp) * + q * / q * = h e i g h t j ORp)* +q*/q*. Let

* * p*/q*
p e Spec(Rp) with minimal over, and having the same height as
 107

J0tp) +q /q . As I ~ J, we a l s o h a v e p minimal over I(Rp) +q . Thus

p e A,(I(Rp) ) by Le~nma 1 . 1 3 , and so by Ler~a 11.32, p NRp e A , ( I R p ) . If

P NR =P , then I c p and P e A,(I)__ as d e s i r e d .

P P --

PROPOSITION 11o35o Let R c T be a faithfully flat extension of Noetherian rings,

and let I he an ideal of R. Then P e A.(1) if and only if there is a

P' e A.(IT) with P' N R = P .

Proof: Lenmaa 11.32 gives one direction, Thus suppose P e A.(1)° Let S = R - P,

so that PS e A.(Is). By Proposition 11.31 there is an integer k such that for

all m > O, ~S : <Ps > ~ PS

"k Now TS is a faithfully flat extension of RS , and

so ~S : <PsTs > ~ P~Ts " By Lermma 11.34, there is a P' e Spec T such that

PS' e A.(IsT S) and PsTs ~ PS' " Of course P' e A.(IT), and since P c P' and

P'Ns=~, we have P'NR=P.

DEFINITION. The sequence of elements x I ,..., x n in the Noetherian ring R is a

strong asymptotic sequence if (xI ,..., Xn) # R and for each i= l,...,n

x i ~U[e e A.((x I ,..., Xi.l)).

Remarks: (a) xI is a strong asymptotic sequence if and only if xI is regular

(A.(0) = Ass R).

~) If (xI,..., Xn) A s = @ with S multiplicatively closed, x I,..., x n a

strong asymptotic sequence in R implies it is one in RS °

(c) xI ..., x is a strong asymptotic sequence in a local ring if and only if it

' n

is one modulo any Q e Ass R.

(d) R-sequences are strong asymptotic sequences (using A.(1) ~ A (I) and the fact

that A (xI ,..., Xn.l) = A s s R/(x I ,..., Xn_l) if x I, ..., Xn_ I is an R-sequence).

(e) Strong asymptotic sequences are asymptotic sequences (roughly speaking, to be

asymptotic, a sequence must behave well modulo any minimal prime in (~)* for

any prime containing it. However strong asymptotic sequences behave well modulo

any prime divisor of zero).

 108

(f) If R c T
_ is a faithfully flat extension of Noetherian rings, x I,o.o, x n in

R is a strong asymptotic sequence in R if and only if it is one in T.

(Proposition 11.35.)

(g) Repeating, we do not know if this definition of strong asymptotic sequence co-

incides with that at the end of Chapter X. It does if n = i or 2.

PROPOSITION 11.36. Let ! be an ideal in a Noetherian ring R and let x I,.,., x n

be a strong asymptotic sequence, maximal with respect to coming from I. Then

n = m i n [ d e p th q*lq* e A s s ( b ) * for I ~ P e Spec R} =min[depth q*lq* e Ass(Rp)* for

P e A,(I)}.

Proof: As the images of x I ,..~ , x n form a strong asymptotic sequence in (~) /q

* , * + q*/q*
for any P ~_ I and q e Ass(Rp) , we easily see height (Xl,.. , Xn)(Rp) = n.

Thus n is equal to or less than the minimum of the first set above, which in turn

is clearly equal to or less than the minimum of the second set. Finally, the maxi-

mality of our sequence shows there is a P e A,(x I, ..o, x n) with I ~ P. By Propo-

sition ii.31, there is a q e Ass(Rp) with (Pp) minimal over

, @ * * * *
(Xl,... , Xn)(Rp) + q . Clearly depth q =height (Pp) /q = n . However (Pp) is

obvionsly also minimal over I(Rp) +q so that by Proposition 11.31, P e A,(I).

Thus the minimum of the second set is equal to or less than n.

Proposition 11.36 shows the following definition is unambiguous.

DEFINITION. For I an ideal in a Noetherian ring, let the strong asymptotic grade

of I, gr, I, be the length of a strong asymptotic sequence maximal with respect to

coming from I.

We now state some results without proof. They are not hard, using the forgoing

machinery.

PROPOSITION II.37. gr I <__ gr. I <_ gr I.

PROPOS ITION ii. 38, If I is an ideal in a local ring (R,M), then gr.l =

gr, IR * =min[height IR* +q*/q*lq* e Ass R*}

 109

PROPOSITION 11.39. The following are equivalent for a Noetherian ring R.

i) gr, l = height I for all ideals I.

ii) g r , M = height M for all maximal ideals M.

iii) R is locally unmixed (i.e. depth q = dim for every P e Spec R,

q e Ass(~) ).

If (R,M) is a local ring and P c Spec R, with S = R -P, then (R)S and

(Rs) do n o t a p p e a r to be closely related. Therefore the next result probably re-

quires strong asymptotic: sequences, or something similar. It can be proved by show-

ing both numbers involved equal gr,P S .

PROPOSITION 11.40. Let P e Spec R, with R local. Then n=m where

p*
n=min{depth q*lq*' e Ass~p)*}t" and m =min[height P*/Q*IQ*' ~ are primes in R ,

Q e Ass R , and P n R = P } .

The analogue of Proposition II.40, which talks about minimal primes rather than
,
associated primes of zero is also true. Its proof uses gr PS '
 APPENDIX: Chain Conditions

This brief appendix states facts referred to in the preceding text, with enough

references that the interested reader may pursue the matter. One new result is pre-

sented here.

DEFINITIONS. (All rings will be Noetherian)

i) A chain of primes P0 ~ P1 c...c Pn is saturated if height Pi+i/Pi= I

for i = 0,1,...,n-l.

li) A saturated chain P0 ~...c p is a maximal chain if P0 and P

n n
are respectively minimal and maximal primes.

lii) The ring R is catenary if any two saturated chains of primes with

common end points have corm~on length.

iv) The local ring (R,M) is quasi-unmixed if every minimal prime in the
*
completion R has depth equal to dim R.

v) The domain R satisfies the Altitude Formula if for any finitely

generated extension domain of R, T, and for any Q e spec T with

P=Q~R, we have height P + T r D ( T / R ) = h e i g h t Q+TrD((T/Q)/(R/P)).

Here TrD(B/A) is the transcendence degree of the domain B over

the subdomain A.

Remark: It is well known that the Altitude Inequality holds for any Noetherian

domain. That is, with R, T, P and Q as above, height P + T r D ( T / R ) >__height Q +

TrD ((T/Q) / (R/P)) [ZS ].

It Was long known that affine rings are catenary. As a consequence of his

structure theorem for complete local rings, Cohen showed that complete local domains

are catenary. Nagata's celebrated example IN , Example 2, pp° 203-205] gives a

Noetherian domain R which is catenary but for which R[X], X an indeterminate, is

not catenary. The following pair of results are fundamental (see [RI ] or [MD]).

THEOREM AI° Let R be a Noetherian ring, and let X be an indeterminate. The

following are equivalent.

 111

i) Every finitely generated extension of R is catenary.

ii) for each maximal ideal M of R, R[X](M,X ) is catenary.

iii) R is locally quasi-unmixed.

iv) ~ is quasi-unmixed for each maximal ideal M of R.

THEOREM A2. Let R be a Noetherian domain. Then R satisfies the Altitude Formula

if and only if R is locally quasi-unmixed.

We note that being quasi-unmixed is quite stable.

THEOREM A3. If the local ring (R,M) is quasi-unmixed, so is every finitely gen-

erated extension. If I is an ideal of R and if all minimal primes of I have

the same depth, then R/I is quasi-unmixed.

Theorem AI ((ii)~=>(iv)) has been generalized (see [MD] or [RM]).

THEOREM A4. Let (R,M) be a local domain and let X be an indeterminate. Let

N e spec R[X] with NnR=M but N#MR[X]. Then the following three sets of in-

tegers are all equal.

a) In I there is a maximal chain of prime ideals of length n in R[X] N}

b) {n I there is a maximal chain of prime ideals of length n-I in some integral

extension domain of R}.

c) {n I there is a minimal prime of depth n-I in R*).

If R c T is an integral extension with (R,M) local, and if T contains a

height 1 maximal then by going up to T and then going down to R, we see that

contains a height i maximal. Thus Proposition 3.19 ((i)<==>(iv)) is a special ease

of Theorem A4.

The next result is referred to in Chapter 5 and Chapter i0. As it does not

appear in the literature, we prove it.

 112

9¢
THEOREM A5. Let P be a prime in a local ring (R,M). Suppose that (R/P) has

a depth n m i n i m a l p r i m e and (Rp) has a depth m minimal prime. Then R has

a depth n + m minimal prime.

Proof: By Theorem A4, there is a maximal chain of length n + l in (R/P)[X]

localized at (M/P,X). Under the natural map R[X] ~" (R/P)[X], suppose the inverse

image of that chain is PR[X] c PI c...c p c ~,X). Now this chain is saturated,
n
and by [HM, Corollary 1.5] we may assume that PINR=P (but PI#PR[X]).

Considering R + ~ + (~) , we let q be a depth m minimal prime of (~)

and q be its inverse image in R. Thus q*/q(~)* is a depth m minimal prime

in (~/q~)*= ((R/q)p/q)*. By Theorem A4. applied to (R/q)e/q with N=PI~[X]/

q~[X], we easily find a saturated chain of primes of length m + l having the form

qR[X] c QI c . . . C Q m c PI "

Clearly qR[X] c QI c...c Qm c PI c P2 c...c Pn c 0~,X) is a maximal chain of

length m + n + l . Applying Theorem A4 to R[X] (M,X) ' we see that R contains a

depth m + n minimal prime.

 REFERENCES

[BI] M. Brodmann, "Asymptotic stability of Ass~/InM) '', proc. Am. Math. Soc.,
74(1979), 16-18.

[B21 , "Asymptotic nature of analytic spreads", Math. Proc. Camb.

Phil. Soc., 86(1979), 35-39.

[B3] "Uber de Minimal Dimension der Assozuerten Primeideale der

Kompletion eines Lokalen Integritatsbereiches", Comment. Math. Helv.,
50(1975), 219-232.

[B4] "Piecewise catenarian and going between rings 't, Pacific J.

Math., 86(1980), 415-419.

[BR] M. Brodmann and C~ Rotthaus, "Local domains with bad sets of formal prime
divisors", J. Algebra, 75(1982), 386-394~

[Bn] L. Burch, "Codimension and analytic spread", Proc. Camb. Phil. Soc.,
72(1972), 369-373.

[D] E. Davis, "Ideals of the principal class, R-sequences~ and a certain

monoidal transformation", Pacific Math. J., 20(1967), 197-205.

[El E.G. Evans, "A generalization of Zariski's Main Theorem", Proc. Am. Math.
Soc., 26(1970), 45-48.

[ES] P. Eakin and A. Sathaye, "Prestable ideals", J. Algebra, 41(1976), 439-454.

[FR] D. Ferrand and M. Raynaud, "Fibres formelles d'un anneau local Noetherian",
Ann. Sci. Ecole Norm Sup., 3(1970), 295-311.

IN1] R. Heitmann, "Prime ideal posets in Noetherian rings", Rocky Mountain J.

Math., 7(1977), 667-673.

[H2] "A non-catenary normal local domain", Rocky Mountain J. Math.,

12 (1982), "~45-148.

E.G. Houston and S. McAdam, "Rank in Noetherian rings", J. Algebra,

37 (1975), 64-73.

[KI] I. Kaplansky, Commutative Rings, University of Chicago Press, 1974.

[K2] , "Adjacent prime ideals", J. Algebra, 20(1972), 94-97.

[Kzl] D. Katz, "Asymptotic primes and applications", Ph.D. Dissertation, Uni-

versity of Texas at Austin, 1982.

[Kz2] , "A note on asymptotic prime sequences", Proc. Am. Math. Soc.,
(to appear).

[Kr] W. Krull, "Zum Dimensionsbegriff der Idealtheorie" (Beitragle zur Arith-

metik Kommutativer Integritatsbereiche, III), Math. Z., 42(1937), 745-766.

[M1] S. McAdam, "l-going down", J. London Math. Soc., 8(1974), 674-680.

[S2] , "Saturated chains in Noetherian rings", Indiana Univ. Math. J.,

23(1974), 719-728.
 114

[M3 ] , "Asymptotic prime divisors and going down", Pacific J. Math.,

91(1980), 179-186.

[M4 ] , "Asymptotic prime divisors and analytic spreads", Proc. Am.

Math. Soc., 80(1980), 555-559.

[~D] S. McAdam and E. Davis, "Prime divisors and saturated chains", Indiana
Univ. Math. J., 26(1977), 653-662.

[ME] S. McAdam and P. Eakin, "The asymptotic ass", J. Algebra, 61(1979), 71-81.

[Ma] J. Matijevic, "Maximal ideal transforms of Noetherian rings", Proc. Am.

Math. Soc., 54(1976), 49-51.

[Mt] H. Matsamura, Commutative Alsebra, Benjamin, 1970.

IN] M. Nagata, Local Rings, Interscience, 1962.

[Ni] J. Nishimura, "On ideal transforms of Noetherian rings II", J. Math. Kyoto
Univ., 20(1980), 149-154.

[ol T. Ogoma, "Non-catenary pseudo-geometric normal rings", Japan J. Math.,

6(1980), 147-163.

JR1] L.J. Ratliff, Jr., "On quasi-unmixed local domains, the altitude formula,
and the chain condition for prime ideals (I)", Amer. J. Math.~ 91(1969),
508-528.

[R2] , "On quasi-unmixed local domains, the altitude formula, and the
chain condition for prime ideals (II)", Amer. J. Math., 92(1970), 99-144.
n
[R3] , "On prime divisors of I , n large", Michisan Math. J.,
23(1976), 337-352.

[R4] , "Two theorems on the prime divisors of zero in completions

of local domains", Pacific J. Math.~ 81(1979), 537-545.

[R5] , "A(X) and GB-Noetherian rings", Rocky Mountain J. Math.,

9(1979), 337-353.

[R6] , "Integrally closed ideals and asymptotic prime divisors",

Pacific J. Math., 91(1980), 445-456.

[R7] , "Note on asymptotic prime divisors, analytic spreads and the

altitude formula", P roc. Am. Math. Soc., 82(1981), 1-6.

[R8] , "On asymptotic prime divisors", Pacific J. Math., (to appear).

[Rg] "Asymptotic sequences", (manuscript).

[RM] L.J. Ratliff, Jr. and S. McAdam, "Maximal chains of prime ideals in in-
tegral extension domains I", Trans. Am. Math. Soc., 224(1976), 103-116.

[RR] L.J. Ratliff, Jr. and D. Rush, "Two notes on reductions of ideals",
Indiana Univ. Math~. J., 27(1978), 929-934.

[Rsl] D. Rees, "Valuations associated with ideals II", J. London Math. Soc.,
36(1956), 221-228.

[Rs2] , "Rings associated with ideals and analytic spreads", Math.

Proc. Camb. Phil. S0c. , 89(1981), 423-432.
 115

[Sl] Jo Sally, "Bounds on generators of Cohen-Macaulay ideals", Pacific Math.

63(1976), 517-520.

[$2] , "A note on integral closure", (manuscript).

[sc] P. Schenzel, "Independent elements, unmixedness theorems, and asymptotic

prime divisors", (manuscript).

[sin] P. Samuel, "Some asymptotic properties of powers of ideals", Annals of

Math.~ 56(1952), 11-21.

[so] M. Sakuma and H. Okuyama, "On a criterion for analytically unramification

of a local ring", J. Gaku~el, Tokushima Univ., 15(1966), 36-38.

[w] K. Whittington, "Prime divisors and the altitude formula", Ph.D. Disser-
tation, University of Texas at Austin, 1980.

[zs] O. Zariski and P. Samuel, Corm~utative Algebra, vol. II, D. Van Nostrand,
1980.
 LIST OF NOTATION

(Page numbers indicate where more information can be found.)

Pase

A(l,n) ASS ~ / I n) 3

A (I) limit of A(l,n), n = l , 2 , ... ,

~(I, n) Ass (R/I n )

A (I) limit of ~(l,n), n=1,2,3,... , 12

B (I,n) Ass (In-l/I)

B (I) limit of B(l,n), n = 1,2,... , 3

F(I,n) Ass (In-1/I n)

B (I) limit of ~(l,n), n=1,2,3,... , I00

gr (I) the asymptotic grade of I 35

the integral closure of the ideal I 3

l~J projective equivalence 53

(i) the analytic spread of I 26

Q(R) the total quotient ring of R

the integral closure of R in Q(R)

R the M-adie completion of the local ring ~,M)

R [1] {yeQ~)l(R : y) ~ P whenever z(P) = I] 77

R<l> [yeQ(R) I(R : y) ! P whenever Ass((Kp)*) 87

contains a depth i prime]
T (I) the ideal transform of I 76

Vi(x) n if x e I n - I n+l; ~ if x e N In , n = 1,2,3,...

lim Vl (xn)/n 39
n->o~
z (P) rain[depth q*Iq* is a minimal prime of (~)*}

C C inclusion , proper inclusion

 INDEX

A I
Altitude Formula 17,41,61,]I0 l-Adic Topology 81
Altitude Inequality II0 Ideal Transform 76
Analytically Independent 63,65 Integral Closure
Analytic Spread 26 of an ideal 3
Artin-Rees Lermma 80,82 of a domain 4,29
Asymptotic Sequence 32
Asymptotic Sequence Over An Ideal 42,58
J
Jacobson Radical 38, 39
B
Brodmann, M. 1,2,51,55,57,87
Burch, L. 55,59 K
Katz, D. 15,32,42,55,57,100
Kaplansky, I. 93
C Krull~ W° 93
Catenary II0
Catenicity 71,72
Chain Conjecture 93,96 L
Cohen-MacAuley 30,35,67 Little Depth 48
Conforming Relation 70 Locally Quasi-Unmixed 28,36,37,53,
Conforming Pair 69,94 58,111

D M
Directly Above 69 Maximal Chain 110,II!
D.V.R. 12,24 Minimal Reduction 26
Monotone Sequence 2,13,15
Multiplicity 30
E
Eakin, P. 9, 30
Evans, E.G. 12 N
Nagata, M. 1 |0
Nishimura, J. 77,79,80
F Normal Domain 30, 63
Flat Extension 76, 84
Faithfully Flat Extension 17,84
0
O gom~s Example 65
G
m

GB (Going Between)-Ring 93, 94

G-Ideal 73 P
Going Down 23,68,73 Permutation 50
Grade 5,38,55,57,59,60 Peskine 12
Graded Ring i Prenormal 96
Prestable 31
Primary Decomposition 9
H Principal Class 28,37,44,67
Heitmann, R. 95 Projective Equivalence 53,76,89
Hilbert Polynomial 26
Homogeneous Graded Ring

Quasi-Unmixed 17,26, 33, 36,44, 49, 64,

II0
 118

index (continued)

R
Ratliff, L.J. Jr. 1,6,7,12,20,32,48,61,79,93
Reduction 14
Rees, D. 12,27,30,32,42,89
Rees Ring 7,15,42
R-Sequence 5,32,35,37,40,52,55,57
Rush, D. 61

S
Samuel, P. 53
Saturated Chain l]O
Sathaye, A. 2
Strong Asymptotic Sequence 88
System of Parameters 28,36

T
Transcendence Degree 12, I ]0

U
UFD 66
Unmixed Local Ring 76
Unmixedness Theorem 40

V
Valuation 90, 91
Vector Space 8

W
Whittington, K. 22,41

Z
n

Zariski's Main Theorem 12

Z-Catenicity 96