Migration velocity estimates have been determined at each position alo
ng the crypt length for both the small and large intestine of the mous
e at 6 different times of the day. Measurements also have been made of
crypt circumference and length. Dramatic, and significant (P<0.001),
changes in migration velocity as a function of time of day were observ
ed in the small intestine with a maximum 0.84 cell positions (cp) per
hour at 0900 h and a minimum of -0.46 cp/h at 1700 h, although the neg
ative velocity was probably artefactual. The 24-h mean velocity rose s
moothly as a function of cell position to a peak of 0.45 cp/h at cell
position 17 (around the top of the proliferative zone). Much more mode
st changes were seen in the percent of (3)HTdR labelled cells (minimum
30.8%, maximum 38.3%, P<0.001) and crypt circumference (minimum 16.9
cells, maximum 17.9 cells, P = 0.003). The migration velocity was rath
er less well determined in the large intestine with a peak in the 24-h
mean velocity (0.26 cp/h) occurring at cell position 10. At this posi
tion significant circadian variation was detected (minimum -0.39 cp/h,
maximum 0.75 cp/h, P = 0.006). Changes were seen in the percent of la
belled cells (minimum 9.4%, maximum 22.3%, P <0.001) and crypt circumf
erence (minimum 18.3 cells, maximum 19.2 cells, P<0.001). In both tiss
ues it is suggested that the combination of the modest changes in cell
proliferation rates in conjunction with the changes in crypt cell num
ber can account for the large amplitude in variation of crypt output,
and that the reservoir effects of changes in crypt geometry are an ess
ential part of the process governing the maintenance of intestinal cel
l numbers.