The objective of this study was to monitor and compare follicle popula
tions and follicular development in pregnant and nonpregnant sows from
Day 3 to Day 20 after breeding. Twenty-four sows were paired within p
arity on the day of artificial insemination and were randomly allocate
d within pair for insemination with either killed (n=12) or live sperm
atozoa (n=12). All the sows were artificially inseminated with the poo
led ejaculate of the same boar. From Day 3 through Day 20 post estrus,
ovarian follicles were scanned daily by ultrasonography. Ultrasound i
mages were recorded on videotape and were retrospectively analyzed. Fo
llicles were mapped to identify the existence of follicular waves. The
follicles were then classified as small (< 3 mm), medium (3-5 mm), or
large (greater than or equal to 5 mm). Pregnancy diagnosis was perfor
med on Day 21 by ultrasonography. Pregnant sows maintained a constant
proportion of the follicle population in the small, medium and large f
ollicle categories. However, in the nonpregnant sows, the proportion o
f follicles in the various size categories remained constant until Day
15. Thereafter, the proportion of small follicles decreased (P < 0.05
) from Day 15 to 20, and the proportions of medium and large follicles
increased (P < 0.05). The predictability of pregnancy status on Day 2
0 based on follicle populations in any of the 3 follicle categories wa
s low. Moreover, there was no evidence of follicular waves during the
estrous cycle or early pregnancy. In conclusion, the proportion of sma
ll follicles decreased while medium and large follicle increased from
Day 15 through Day 20 of the estrous cycle, but not during a similar s
tage of pregnancy. This latter finding concurs with follicle recruitme
nt from the pool of small follicles for ovulation following PGF2 alpha
secretion to induce luteolysis, which reduces progesterone concentrat
ions and thereby allows for the stimulation of the pool of small folli
cles by gonadotropins.