Controlling First Service and Calving Interval by Prostaglandin F2e, Gonadotropin-Releasing Hormone, and Timed Insemination 1

M A T T H E W C. LUCY, 2 JE FFREY S. STEVENSON, 3 and E. P. CALL Department of Animal Sciences and Industry

Kansas State University Manhattan 66506

A B S T R A C T

Our objective was to determine if calving intervals could be shortened and made less variable by using prostaglandin F2~ to control the occurrence of first services. Holstein cows (n=348) were assigned at calving to four t reatment groups. Control cows (n=88) were insemi- nated at their first observed estrus after 40 d postpartum. Estrous cycles of the remaining cows were synchronized with prostaglandin F2a to allow insemination (first services) 80 h after the second injec- t ion (n=86), insemination at 80 h pre- ceded by gonadotropin-releasing hormone at 72 h (n=86), or insemination at 72 and 96 h (n=88) after the second injection (51 to 57 d postpartum). By design, inter- val to first service was reduced to 57 d for t reated cows (63 d for controls) and was less variable (12% of that for controls). Conception rate at first service was lower after t imed inseminations than that of controls. Intervals to conception and sub- sequent calving were similar in all treated cows and controls. Of cows sampled, 23 of 176 (13%) failed to respond with luteolysis when progesterone in serum exceeded 1 ng/ml, and 26 of 176 (15%) had low concentrations of progesterone in serum and could not respond to pros- taglandin F2~. Poor response to t imed inseminations may have occurred because only 72% of 176 cows sampled responded with luteolysis. When only cows observed in estrus were considered, conception

Received February 7, 1986. 1Contribution Number 86-283-J, Kansas Agricul-

tural Experiment Station. Department of Animal Science, North Carolina

State University, Raleigh. 3 Reprint requests.

rate approached that of controls (51%). Methods of t imed insemination utilized and their resulting low conception rates prevented a true test of the uti l i ty of estrous synchronization to reduce dura- t ion and variability of calving intervals, even though variability of intervals to first service was reduced.

I N T R O D U C T I O N

Calving interval has profound economic effects in the dairy herd (5, 14). Numerous studies have demonstrated the benefit of a 12 to 13-too calving interval (1, 5, 8, 14, 19). In te~als longer than opt imum result when cows spend a greater proport ion of their lacta- t ion in the less profitable part of their lactation curve (5). Thus, economic losses associated with prolonged calving intervals often go un- detected by dairy producers, because they represent lost potential income and not cash outlay (5). Estrous detection and interval to first service are the most important factors affecting calving interval (15). Poor estrous detection has been linked to reproductive inefficiency in many studies, and its effect on calving interval is well-documented (2, 3, 15). Interval to first service has received less at tent ion than estrous detection as a measure of reproductive performance. Furthermore, calving interval and interval to first service are highly correlated (4, 10, 18), and optimal calving intervals may be unattainable without decreasing intervals to first service (5).

Unfortunately, reducing calving intervals to an optimal average on a herd basis does not optimize necessarily individual calving intervals for all cows in the herd. The herd average can include cows with extremely short or extended intervals. Reducing interval to first services will shorten calving intervals (10, 20), but postpar- tum inseminations earlier than 60 d may lower conception rates (4). An option through estrous

1986 J Dairy Sci 69:2186-2194 21 86

CONTROLLING FIRST SERVICE AND CALVING INTERVAL 2187

synchronization is to schedule all first services at approximately 60 d in an a t tempt to shorten the duration and variability of calving intervals. Our objectives were 1) to determine if duration and variability of calving intervals could be reduced by controlling time of first services with prostaglandin F2a (PGF); 2) to evaluate the reproductive performance of dairy cows subjected to three treatments using t imed insemination after PGF and gonadotropin- releasing hormone (GnRH); and 3) to assess efficacy of PGF-induced luteolysis and estrus and its subsequent effect on the success of t imed inseminations.

MATERIALS AND METHODS

Experimental Design

Lactating Holstein cows (n=348) from the Kansas State University dairy herd that calved between July 1, 1983 and June 30, 1985 were utilized. Cows were housed in a freestall con- f inement facility exposed to the environment. The diet consisted of a concentrate mix (16% protein) containing 50% corn and 50% sorghum grains, soybean meal, 1.5% sodium bicarbonate, and minerals in a self-feeder and alfalfa hay fed ad libitum. Cows were milked twice daily at 0930 and 2130 h. Two daily 30-min observa- tions (0700 to 0900 h and 1600 to 1800 h) were conducted for estrous detection. At calving, cows were assigned to four t reatment groups. Control cows (n=88) were inseminated artificially (AI) at first observed estrus after 40 d postpartum. Time of first services was scheduled in the remaining cows using two 25-mg i.m. injections of PGF (PGF-1 and PGF-2; Lutalyse®; The Upjohn Company, Kalamazoo, MI) 11 d apart. Small groups of cows were treated together on Thursdays of each week (0800 to 0900 h) when cows were 40 to 46 d postpartum. Eleven days later (Mondays), cows were given a second injection (PGF-2) between 0800 and 0900 h. Following PGF-2 (0 h), cows were treated according to the following procedures: 1) AI at 80 h (n=86); 2) 100 /ag (i.m.) GnRH (Cystorelin ®, CEVA Laboratories, Overland Park, KS) at 72 h and A1 at 80 h (n=86); or 3) AI at 72 and 96 h (n=88). Injections of GnRH occurred on Thursdays (0800 to 0900 h), 80-h insemina- tions on Thursday afternoons (1600 to 1700

h), and 72 and 96-h inseminations on Thursday and Friday mornings (0800 to 0900 h). Insemi- nations at 72 and 96 h were counted as one service. Cows were reinseminated according to observed estrus until determined pregnant by uterine palpation per rectum at 40 to 46 d after AI or removed from the herd for various rea- sons including infertility.

Blood Collection

Blood was collected by coccygeal venipunc- ture from a random group of PGF-treated cows (n=176) at 0 and 48 h after PGF-1 and PGF-2. Samples were stored at 5°C for 24 h until serum was harvested by centrifugation. Samples of serum were held at - 2 0 ° C until assayed for progesterone (P) by radioimmunoassay (21). lntraassay a n d interassay coefficients of varia- t ion were 10.3 and 15.1%, respectively. A luteolytic response after PGF-2 was indicated when P in serum exceeded 1 ng/ml at 0 h and was less than 1 ng/ml at 48 h. No luteolysis was indicated when P was greater than 1 ng/ml at 0 and 48 h after PGF. Serum P of less than 1 ng/ml at 0 and 48 h suggested that cows did not have a functional corpus luteum.

Statistical Analyses

A different inseminator was used during the first and second year of the study. Because no effect of inseminator was found, data were combined. Intervals to first service, first service to conception, conception rate at the first three services, calving interval (estimated by adding 280 days to the interval from calving to conception), and services per conception and per cow were analyzed by the least squares procedures of the Statistical Analysis System (17). Treatment (n=4), lactation (primiparous or muhiparous), season of calving (October 1 to January 31, February 1 to May 30, or June 1 to September 30), and all two-way interactions were included as sources of varia- tion. A measure of the variation associated with the mean intervals to first service, to first service conception, and to subsequent calvings was determined by Levene's test (12). This was accomplished by calculating a residual for each observation (absolute value of the difference between the observed value and treatment mean), which was subjected to analysis of vari- ance. When the resulting F-test was significant,

Journal of Dairy Science Vol. 69, No. 8, 1986

2188 LUCY ET AL.

variances were heterogeneous, and synchrony of fit for individual observations about the mean were compared by the relative magni- tudes of the residual means for each treatment. Smaller mean residuals indicated less variability within treatment. Orthogonal contrasts were made between treatment means (control vs. timed AI groups, timed AI with and without GnRH, and one vs. two timed inseminations). Selected nonorthogonal contrasts were made by Scheffe's interval (17). Enumeration data (percentages) were tested for independence by chi-square.

R E SU L TS

Intervals from Parturition to First Service and Conception

Analyses of variance for various reproduc- tive traits are in Table 1. Interval to first service was affected only by treatment as average inter- vals to first insemination were 6 d shorter (P<.001) for PGF-treated cows than controls (Table 2). By design, timed AI after PGF re- sulted in less variable intervals to first service, since the mean residuals for treated cows were 12% of those for controls. The range of first services was reduced to 7 d (54 to 60 d post- partum) compared with 83 d (40 to 122 d) for controls.

Interval from first service to conception was altered by treatment (P<.05) and season of calving (P<.05). Treatment (P<.01) and season (P<.001) accounted for some of the variation in the mean residuals for this interval (Table 1). Cows inseminated at 72 and 96 h and those in- seminated at 80 h that also received GnRH had longer (P<.01) and more variable (P<.01) inter- vals to conception after first service than cows inseminated at 80 h without GnRH treatment (Table 2). However, AI at 80 h without GnRH treatment resulted in less variable intervals from first service to conception than those of con- trols (P<.05) and the remaining AI groups (P<.01). Cows calving in the spring had longer (50 -+ 6 d; P<.O1) and more variable (P<.O01) intervals from first service to conception than cows calving in fall and winter (31 + 4 d), whereas cows calving in summer had interme- diate intervals (40 + 6 d).

Interval between first and second service averaged 36 +- 4 d for control cows, whereas that interval for all PGF-treated cows averaged

Journal of Dairy Science Vol. 69, No. 8, 1986

34 + 3 d and was similar to controls. Intervals from second to third service also were similar for controls (36 +- 4 d) and PGF-treated cows (40 + 3 d). However, the proportion of PGF- treated cows with intervals from treatment AI (first service) to second service of less than 18 d (33/181; 18%) was greater (P<.01) than that for all control (intervals from first to second service for control cows and those from second to third services for all cows) cycles (3/155; 2%).

Calving intervals were affected by treat- ment (P<.10) and season (P<.05) as were the residuals of those factors (Table 1). Although calving intervals were not lengthened signifi- cantly by timed insemination treatments com- pared with controls, GnRH-treated cows had longer (P<.01) calving intervals than their coun- terparts inseminated at 80 h (Table 2). As with the intervals to conception after first service, variation in calving interval was not reduced for treated cows, but variation was less (P<.05) for cows inseminated at 80 h (without GnRH) than for controls and for the remaining treated groups that were first inseminated by appoint- ment. Cows calving during fall and winter sea- sons had shorter (370 +- 4 d; P<.05) and less variable (P<.001) calving intervals than those in spring (388 +- 6 d), whereas summer calvings were intermediate (378 -+ 6 d).

Conception Rates and Fertility

Conception rate at first service (treatment) was altered (P<.01) only by treatment (Table 1). Control cows had higher (P<.01) concep- tion rates at first service (51%) than all timed AI groups (Table 3). Conception rate at first service was similar among all timed A1 groups. Detection rate of estrus after PGF-2 was simi- lar among treatments (Table 3). When only cows that were observed in estrus after PGF-2 were considered for determining conception rate at first service, conception rates of estrual cows were higher than those of all treated cows [33% (19/58), 41% (23/56), and 37% (23/63) for cows inseminated at 80 h, at 80 h plus GnRH, and at 72 and 96 h, respectively], but were still less (P<.05) than those for controls.

Rates of conception at second and third services were similar for PGF-treated cows and controls, except that cows previously treated with GnRH and inseminated at 80 h had a lower (P<.05) conception rate at second ser-

T A B L E 1. L e a s t s q u a r e s a n a l y s e s o f v a r i a n c e o f i n d i v i d u a l o b s e r v a t i o n s a n d t h e i r r e s i d u a l s f o r r e p r o d u c t i v e t r a i t s o f d a i r y c o w s a f t e r t i m e d i n s e m i n a t i o n s in c o w s t r e a t e d w i t h p r o s t a g i a n d i n F 2 a ( P G F ) a n d g o n a d o t r o p i n - r e l e a s i n g h o r m o n e ( G n R H ) .

S e r v i c e s

D a y s F i r s t p e r t o f i r s t s e r v i c e t o C a l v i n g C o n c e p t i o n r a t e C o n -

S o u r c e o f v a r i a t i o n d f s e r v i c e c o n c e p t i o n i n t e r v a l 1 s t 2 n d 3 r d c e p t i o n C o w

(3 0 Z

0

e~

T r e a t m e n t 3 * * , 2 , , ,3 , 2 , , 3 , 2 , ,3 , , 2 , 2 , 2 **2 C o n t r o l vs . t i m e d i n s e m i n a t i o n 1 * * * * * * * * * * ** G n R H vs. n o G n R H 1 ** * * * ** * * * ** * f O n e vs. t w o i n s e m i n a t i o n s 1

S e a s o n 2 * * * * * * * * ~" * * F a l l - w i n t e r vs . s p r i n g 1 * * * * * * * * * * * * * F a l l - w i n t e r vs . s u m m e r 1 * t"

L a c t a t i o n n u m b e r 1 T r e a t m e n t × s e a s o n 6 T r e a t m e n t × l a c t a t i o n 3 L a c t a t i o n × s e a s o n 2

Z

<

t~

Z

c3 > t~

Z

e~

<

Z o o0

ox

i T h e G n R H ( 1 0 0 / z g ) w a s g i v e n 72 h a f t e r P G F .

2 S i g n i f i c a n c e f o r i n d i v i d u a l o b s e r v a t i o n s .

3 S i g n i f i c a n c e f o r s y n c h r o n y o f f i t o f i n d i v i d u a l o b s e r v a t i o n s a b o u t t h e i r m e a n ( L e v e n e ' s t e s t ) . S e e M a t e r i a l s a n d M e t h o d s f o r f u r t h e r de t a i l s .

* P < . 0 5 .

* * P < . 0 1 .

* * * P < . O 0 1 .

t P < . l O .

ro

2 1 9 0 LUCY ET AL.

TABLE 2. Average (X,) and mean residual (Xr) days from parturition to first service, first service to conception, and to subsequent calving after timed inseminations at first service using prostaglandin F2a (PGF) and gonado- tropin-releasing hormone (GnRH).

Timed inseminations 1

80 h + 7 2 +

Item Control 80 h GnRH 96 h

Days to first service

First service to conception

N 88 86 86 88 63 57 a 57 a 57a

Xr 17 2 a 2 a 2a

N 77 68 70 72 33 31 54a, b 44

Xr 38 26 a 45 b 40b

Calvinginterval N 77 68 70 72 375 368 391 b 381

Xr 39 26 a 45 b 41b

aDifferent from control (P<.05).

bDifferent from AI at 80 h (P<.01).

i GnRH (100 #g) was given 72 h after PGF.

TABLE 3. Measures of fertility and reasons for culling after scheduled first services using prostaglandin F2a (PGF) and gonadotropin-releasing hormone (GnRH).

Timed inseminations 1

Item Control 80 h 80 h + GnRH 72 + 96 h

Conception rate First service 45/88 (51) 2 20/86 (23) a 26/86 (30) a 27/88 (31) a Second service 15/36 (42) 32/64 (50) 16/59 (27) b 26/58 (45) Third service 10/21 (48) 12/30 (40) 17/38 (45) 7/32 (22)

Observed estrus after PGF 58/86 (67) 56/86 (65) 63/88 (72) Services/conception 1.9 -+ .1 2.0 -+ .1 a 2.5 +- .2a, b 2.1 -+ .2 a Services/cow 2.0 -+ .2 2.2 +- .2 a 2.7 -+ .2 a 2.6 -+ .2 a Pregnancy rate 77/88 (87) 68/86 (79) 70/86 (81) 72/88 (82)

Reasons for culling Reproduction 5 (6) 3 8 (15) 9 (10) 11 (11) Milk production 4 2 2 2 Mastiffs 1 1 2 1 Ketosis 0 1 0 0 Injury 0 3 2 1 Death 1 3 1 1

a . Different from control (P<.05).

bDifferent from timed AI at 80 h (P<.05).

1GnRH (100 ~g) was given 72 h after PGF.

2 Percent.

Percentage of cows treated.

Journal of Dairy Science Vol. 69, No. 8, 1986

CONTROLLING FIRST SERVICE AND CALVING INTERVAL 2191

vice (27%) than those without GnRH treatment (50%) (Table 3). Cows that calved during spring months had lower (P<.05) conception rates at second service (31%) than those calving during fall and winter months (46%) with summer being intermediate (43%). There were no sea- sonal effects on conception rate at first and third services.

Conception rates did not appear to be related to milk yield of cows because daily milk yield of control cows (29.4 + .7 kg) nearest the day of first service was similar to that for PGF- treated cows (29.6 + .8 kg). Lactation yields (mature equivalent, 305-d, 2X) for controls (7665 + 144 kg) and PGF-treated cows (7673 +- 162 kg) also were similar. As expected, daily milk yield was greater (P<.001) for multiparous (34.1 + .7 kg) than for primiparous cows (24.9 + .8 kg), and both daily and full lactation yields were greater (P<.05) in spring than fall and winter or summer months.

Services per conception (only cows that con- ceived) and services per cow (all cows regardless of their eventual conception status) were af- fected (P<.05) by treatment and season (Table 1). Control cows required fewer (P<.05) ser- vices per conception and per cow than PGF- treated cows (Table 3). Furthermore, cows previously treated with GnRH required more (P<.05) services per conception than those in- seminated at 80 h and not receiving GnRH. Cows calving in fall and winter had fewer (P<.05) services per conception (1.9 -+ .1) and per cow (2.1 + .1) than those calving during spring (2.3 + .1 and 2.6 -+ .1) and summer (2.2 + .1 and 2.4 + .2).

Despite lower conception rates at first ser- vice and more services per conception, preg- nancy rates for control and treated cows were similar (Table 3). Reasons for cow disposal were similar; however, culling for reproductive failure was greater for cows treated with PGF than for control cows (Table 3).

Efficacy of Prostaglandin F2e for Timed Insemination

Based on concentrations of P in serum from 176 cows at 0 and 48 h after each of two injections of PGF (PGF-1 and PGF-2), three different responses were observed. These are illustrated for treated cows after PGF-2 in Figure 1. Progesterone decreased (P<.05) 48 h

after PGF-2 for 127 cows, indicating luteolysis, but in another 23 cows, P in serum was un- changed (no luteolysis). Surprisingly, 26 cows had low concentrations of P in serum at 0 and 48 h after PGF-2 and could not have responded to PGF, despite a previous injection of PGF 11 d earlier (between 40 and 46 d postpartum).

Comparison of the responses after the first and second injections of PGF is in Table 4. Before PGF-1, 123 of 176 cows (70%) had a functional corpus luteum (serum P greater than 1 ng/ml). This percentage increased (P<.01) to 85% 11 d later before PGF-2. Luteolysis oc- curred in 60% of all treated cows after PGF-1 and increased (P<.01) to 72% after PGF-2. However, the percentage of cows with luteo- lysis was identical (85%) after both PGF-1 and PGF-2.

No luteolysis occurred in 10 and 13% of all treated cows after PGF-1 and PGF-2, respec- tively. In other words, 15% of cows with a functional corpus luteum failed to have luteo- lysis after either injection of PGF. Of 23 cows that failed to have luteolysis when serum P was greater than 1 ng/ml at 0 and 48 h after PGF-2 (Figure 1), 16 (70%) responded to PGF-1, 4 (17%) had low serum P at PGF-1, and 3 (13%) failed to respond to PGF-1.

5 5 T [ ] Luteolysis at PGF-2

(n= 127) 72% 4 [ [ ] No luteolysis at PGF-2

1 (n=23) I 3% [ ] Low progesterone of PGF-2

v (n=26) 15%

0 h 48

Time Af te r PGF Figure 1. Mean (-+ SD) serum progesterone concen-

t-rations for dairy cows following the second (PGF-2) of two injections of prostaglandin F ~ (PGF) given 11 d apart. Cows are grouped according to their progesterone response: luteolysis, no luteolysis, or low serum progesterone at 0 and 48 h after PGF.

3 o x_

o 2

£1-

Journal of Dairy Science Vol. 69, No. 8, 1986

2192 LUCY ET AL.

TABLE 4. Responses to prostaglandin F2a (PGF) at 11-d intervals preceding timed inseminations.

Injection of PGF 1 Item PGF-1 PGF-2

No, cows treated 176 176 No. cows

Functional corpus luteum 123 (70) 2 150 (85) a Luteolysls 105 (60) 127 (72) a No luteolysis 18 (10) 23 (13) Low serum progesterone 53 (30) 26 (15) a

aDifferent from proportion of cows at first injec- tion (P<.01).

1 First injection of PGF (25 mg) was given at 40 to 46 d postpartum.

2 Percentage of cows treated.

Fewer (P<.01) cows had low concent ra t ions o f P in serum at 0 and 48 h after PGF-2 than after PGF-1 (Table 4). Of 26 cows with low P at PGF-2 (Figure 1), 9 (35%) responded to PGF-1, 9 (35%) had low P at PGF-1, and 8 (30%) failed to respond to PGF-1 when P in serum exceeded 1 ng/ml.

The p ropor t ion of cows observed in estrus was related to the response after PGF. Fewer (P<.05) cows (27/47 or 57%) were observed in estrus after PGF-2 when cows had low serum P (n=7) or failed to respond to PGF-1 (n=40) than when cows responded (luteolysis) to both inject ions of PGF (60/80 or 75%). Concep t ion rates also t ended to be higher (P<.10) for cows that responded to both PGF inject ions (32 /80 or 40%) than for those only responding to PGF-2 (12/47 or 25%). Of 49 cows that e i ther failed to respond to PGF-2 or had low concen- t ra t ions o f P at 0 and 48 h after PGF-2, one cow conceived. Tha t cow had low P, was no t observed in estrus, and was inseminated at 72 and 96 h after PGF-2.

If only first-service concept ion rates of cows with luteolysis (127/176) are considered by t r ea tmen t group, 10 of 43 (23%) cows insemi- nated at 80 h, 18 of 43 (42%) cows inseminated at 80 h and receiving GnRH, and 16 of 41 (39%) inseminated at 72 and 96 h conceived at first service. Gonadotropin-re leas ing ho rmone tended to increase (P=.07) first-service concep- t ion rates for cows inseminated at 80 h. How-

ever, these percentages were still less (P<.05) than that of controls (51%). However , when cows with luteolysis and observed estrus were considered after PGF-2, concep t ion rate ap- proached that of controls (9 /30 or 30%, 14/30 or 47%, and 15/27 or 55%). Resul t ing calving intervals of these cows were shorter (P<.05) than those for cows that failed to have luteo- lysis or could n o t have responded to PGF be- cause o f low P in serum. These differences were 13, 27, and 43 d, respectively, for the three groups of PGF-t rea ted cows.

DISCUSSION

Timing first services to occur at approxi- mate ly 57 d pos tpa r tum using PGF al lowed eff icient scheduling of inseminations. Al though average interval to first service was within 6 d of that for control cows, variabil i ty was re- duced to 12% of cont ro l intervals. Unfor tu- nately, variability of calving intervals was no t reduced. However, by delaying any first insemi- nat ion of PGF-t rea ted cows to 54 to 57 d post- pa r tum (14 to 17 d longer than controls) , we did not lengthen calving intervals nor delay interval to concept ion. We should poin t ou t that the value of estrous synchroniza t ion as a tool to test our hypothesis was not fully realized because the me thods o f t imed insemi- na t ion resulted in low concept ion rates. A more effect ive approach to improve concept ion rate at first service and possibly reduce variat ion in calving intervals would involve synchronizing estrus and only inseminating cows observed in estrus. A single dose of PGF fo l lowed by in- seminat ion at estrus might produce acceptable results, because percentage of total cows with luteolysis after PGF-1 was 60 and only in- creased to 72 when two inject ions o f PGF were ut i l ized (Table 4). A proven al ternat ive to reduce interval to concept ion is to inseminate all cows at observed estrus and then treat remaining cows (not observed in estrus) wi th PGF, based on detec t ion of a palpable corpus lu t eum (16). The impact of these systems on reducing variabili ty of subsequent calving inter- vals has no t been determined.

Success of our t rea tments depended on pres- ence of a viable corpus lu teum, eff icacy of PGF to regress that corpus lu teum, and the presence of a new corpus lu teum responsive to PGF within 11 d after PGF-1. Abi l i ty of PGF treat-

Journal of Dairy Science Vol. 69, No. 8, 1986

CONTROLLING FIRST SERVICE AND CALVING INTERVAL 2193

ments to control estrus by inducing luteolysis after two PGF injections was poor compared with previous work in heifers (6, 21). Even though 85% of treated cows having a corpus luteum responded with luteolysis after each PGF injection, this was less than the 91% in a previous report (16). However, these results are similar to recent reports for observed estrus in lactating cows after PGF (11).

Another limitation encountered was absence of a functional corpus luteum (low concentra- tions of P in serum) in 26 (15%) of 176 cows at PGF-2. This percentage seems rather high in view of the fact that all treated cows received PGF 11 d earlier and the average postpartum interval to first ovulation for our herd was 19 + 1 d in a previous study (20). These repre- sented, in part, cows that responded to PGF-1 but failed to have serum P that exceeded 1 ng/ml at PGF-2, indicating a viable corpus luteum. Based on daily blood samples in selected cows after PGF-1, we have determined that these cows had a delayed rise in serum P resulting in low P 11 d later at PGF-2. A second reason for low P at the time of PGF-2 was failure of cows to respond to PGF-1. The third reason for low P at PGF-2 was anestrus or anovulation, because 5% (9/176) of the cows were apparently not cycling by 51 to 57 d postpartum.

It is not surprising that insemination of 28% (49/176) of the cows (those that failed to respond to PGF or had low serum P) only resulted in one conception. Insemination of those cows not in estrus reduced overall con- ception rates and probably contributed to our inability to reduce variability of resulting inter- vals to conception and calving. Despite low conception rates at first service, PGF treat- ments did not prolong intervals from calving to conception. Part of the reason that PGF treat- ments did not prolong conception intervals might be due to a greater proportion of PGF cows that were re-inseminated within 18 d after the timed AI. However, those cows with luteo- lysis and observed estrus tended to have im- proved rates of conception and decreased intervals to conception than control cows or cows treated with PGF without luteolysis or estrus.

Compared with their own second service conception rates and those of controls at first service, PGF-treated cows had lower first

service conception rates. Cows previously treated with GnRH also showed poor concep- tion at second service. This may be related to reduced P in serum during the luteal phase after GnRH treatment (13; M. C. Lucy and J. S. Stevenson, 1986, unpublished data). Several studies have suggested a positive relationship between P in serum during the luteal phase pre- ceding AI and the success of that insemination (7, 8). Because P was reduced in our GnRH- treated cows after first service as in another experiment (M. C. Lucy and J. S. Stevenson, 1986, unpublished data), this might account for the low second service conception rates. The resulting low first and second service concep- tion rates for GnRH-treated cows also ex- plained the longer intervals to conception for this treatment group. In contrast, for cows with luteolysis (verification that cows were estrual), GnRH (at 72 h) improved first service concep- tion rates when insemination occurred at 80 h. This increase in conception rate has been reported by others (9). Therefore, low concep- tion rate at second service may result in cows treated with GnRH during a nonestrous period (e.g., luteal phase), because reduced P in serum preceded second service.

Our study demonstrated that average dura- tion and variability of calving intervals were unaffected by our treatments, even though first services did not begin until 54 to 57 d post- partum for PGF-treated cows and variability of those intervals to first service was reduced to 12% of that for controls. Lower conception rate at first service and less than optimal estrous synchrony probably prevented reduction of the mean and variance for calving intervals. Our results also demonstrated that an acceptable calving interval for the control cows (375 d) was achieved with a reasonable twice daily estrous detection program similar to that shown by Allalout (1). Other treatments using estrous synchronization that optimize first service con- ception and also reduce variability of intervals to first service may prove useful in reducing variability and length of calving intervals for all cows in a herd.

ACKNOWLEDGMENTS

The authors express gratitude to M. K. Schmidt and K. D. Frantz for their assistance in conducting this study at the Kansas State

Journal of Dairy Science Vol. 69, No. 8, 1986

2194 LUCY ET AL.

Universi ty Dairy Center , to S. Durham for her assistance in the labora tory , and to B. Beckley fo r typ ing this manuscr ip t . We acknowledge the dona t i on of pros taglandin F2a (Lutalyse ®) by J. F. McAllister o f the Up john Company , Kalamazoo, MI, and the gift o f gonado t rop in - releasing h o r m o n e (Cystorel in ®) by M. D. Brown of CEVA Laborator ies , Overland Park, KS.

REFERENCES

I Allalout, S. 1979. The effect of calving interval, days open, previous days dry, and percentage cow days in milk on annual yield per cow. M.S. Thesis, Washington State Univ., Pullman.

2 Barr, H. L. 1975. Influence of estrous detection on days open in dairy herds. J. Dairy Sci. 58:246.

3 Bozworth, R. W., G. Ward, E. P. Call, and E. R. Bonewitz. 1972. Analysis of factors affecting calving intervals of dairy cows. J. Dairy Sci. 55: 334.

4 Britt, J. H. 1975. Early postpartum breeding in dairy cows: a review. J. Dairy Sci. 58:266.

5 Call, E. P. 1978. Economics associated with calving intervals. Page 190 in Large dairy herd manage- ment. C. J. Wilcox and H. H. Van Horn, ed. Univ. Presses Florida, Gainesville.

6 Cooper, M. J. 1974. Control of estrous cycles of heifers with a synthetic prostaglandin analogue. Vet. Rec. 95:200.

7 Folman, T., M. Rosenberg, Z. Herz, and M. Davidson. 1973. The relationship between plasma progesterone concentration and conception in postpartum dairy cows maintained on two levels of nutrition. J. Reprod. Fertil. 34:267.

8 Fonseca, F. A., J. H. Britt, B. T. McDaniel, J. C. Wilk, and A. H. Rakes. 1982. Reproductive traits of Holsteins and Jerseys. Effects of age, milk yield, and clinical abnormalities on involution of cervix and uterus, ovulation, estrous cycles, detection of estrus, conception rate, and days open. J. Dairy Sci. 66:1128.

9 Hansel, W., and J. Fortune. 1978. The application

of ovulation control. D. B. Crighton, N. B. Haynes, G. R. Foxeroft, and G. E. Lamming, ed. Pages 237-263 in Control of ovulation. Butterworth, London.

10 Harrison, D. S., C. E. Meadows, L. J. Boyd, and J. H. Britt. 1974. Effect of interval to first service on reproduction, lactation and culling in dairy cows. J. Dairy Sci. 57(Suppl.):628. (Abstr.)

11 MacMillan, K. L. 1983. Prostaglandin responses in dairy herd breeding programmes. N.Z. Vet. J. 31: 110.

12 Milliken, G. A., and D. E. Johnson. 1984. Levene's test. Pages 19--25 in Analysis of messy data. Vol. I. Designed experiments. Lifetime Learning Publ., Belmont, CA.

13 Milvae, R. A., B. D. Murphy, andW. Hansel. 1984. Prolongation of the bovine estrous cycle with a gonadotropin-releasing hormone analog. Biol. Re- prod. 31:664.

14 Olds, D.~ T. Cooper, and F. A. Thrift. 1979. Effect of days open on economic aspects of current lacta- tion. J. Dairy Sci. 62:1167.

15 Pelisser, C. L. 1972. Herd breeding problems and their consequences. J. Dairy Sci. 55:385.

16 Plunkett, S. S., J. S. Stevenson, and E. P. Call. 1984. Prostaglandin Fza for lactating dairy cows with palpable corpus luteum but unobserved estrus. J. Dairy Sci. 67:380.

17 Statistical Analysis System. 1979. SAS User's guide. SAS Inst., Inc., Cary, NC.

18 Slama, H., M. E. Wells, G. D. Adams, and R. D. Morrison. 1976. Factors affecting calving interval in dairy herds. J. Dairy Sci. 59:1334.

19 Speicher, J. A., and C. E. Meadows. 1967. Milk production costs associated with length of calving interval on Holstein cows. J. Dairy Sci. 50:975. (Abstr.)

20 Stevenson, J. S., and E. P. Call. 1983. Influence of early estrus, ovulation, and insemination on fer- tility in postpartum Holstein cows. Therlogenology 19:367.

21 Stevenson, J. S., M. K. Schmidt, and E. P. Call. 1984. Stage of estrous cycle, time of insemination, and seasonal effects on estrus and fertility of Hol- stein heifers after prostaglandin F2a. J. Dairy Sci. 67:1798.

Journal of Dairy Science Vol. 69, No. 8, 1986