410 JORDAN AND EYCLESHYMER. [Vor. 1X. 
lower pole. In the largest eggs (Amblystoma) the vertical 
furrows consume nearly two hours in reaching the lower pole, 
while in the eggs of Dzemyctylus but little over an hour is 
needed for the same operation. The furrows in the newt and 
in the frog, on the contrary, advance towards the lower pole 
with about equal velocity, although, as we have stated, the 
inter-cleavage period is much the longer in the newt. 
We have already indicated in a preliminary note the more 
important varieties of cleavage that we observed (Anat. Anz. 
VII, 1892). The proportion of eggs that swerved from the 
Amphibian “type” somewhat surprised us. In a batch of 
seventy-one frog eggs from one mother, for example, the first 
furrow divided all the eggs into nearly equal hemispheres. 
The second furrow cut the first at nearly right angles in the 
median line in sixty-nine eggs. In the other two eggs furrows 
appeared as shown in Fig. 5 ; these two eggs were isolated, and 
produced perfectly normal embryos. The next furrow was truly 
horizontal in only twenty-nine out of the sixty-nine eggs 
remaining ; the other forty showed considerable variation. In 
some of these forty eggs equatorial furrows appeared in three 
quadrants, and a true vertical in the other quadrant (Figs. 2 
and 8). In other eggs three cleavage planes were vertical 
and one horizontal (Fig. 26). In still other eggs the whole 
third set of furrows was vertical (Fig. 33). 
This large proportion of variations from the normal is not 
uncommon, although in other batches of eggs a smaller number 
of variations was usually observed. In this respect there is 
considerable difference between eggs from different parents. 
Eggs from some parents showed far greater tendency to 
vary in the early stages than did others. In general, about 
one half of the eggs fail to form a true “first equatorial” 
plane in all four quadrants. Some of the more interesting of 
these deviations from the type are shown in Plate XXVI. 
As the number of cells increases the cell-divisions become 
less synchronous. Some cells show a tendency to divide more 
rapidly than others, and consequently the furrows do not all 
appear simultaneously, but follow one after the other at per- 
ceptible intervals. In the fifth and sixth set of furrows, for 
