202 University of California Publications in Zoology (Vou.9 
it quite certain that many of them ... . not only are subject to 
definite schemes as to positional arrangement and time of origin, 
but also as to mass relations. Otherwise stated, the repetitive 
series which obviously constitute so large a part of the whole 
animal are to a great extent disposed in mathematically treat- 
, 
able order.’’? Although the systematically graded quantitative 
relation among repeated parts was definitely recognized in this 
study, no attempt was made to measure the quantities. 
The first effort toward dealing with the problem quantita- 
tively was made by Miss Johnson (1910). Her most immediate 
results were that ‘‘the salpa chain presents an obvious period- 
icity,’’ the so-called ‘‘blocks’’ of zooids making the periods; and 
that the zooids at the ends of the block are much smaller than 
the intermediate ones, those of maximum size being nearer the 
distal end in the younger blocks, and nearer the proximal end in 
the older blocks. Attention was called to the resemblance of the 
size-scheme of zooids in the blocks to the well-known phenomenon 
of ‘‘grand period of growth’’ exhibited by many plants. 
Ritter and Johnson (1911) found that in the wheel-shaped 
eroups characterizing the chain of another species of salpa, the 
size gradation of the zooids is more generally distributed in the 
groups than in the species previously studied, the graphs made 
from plotting the measurements approaching more closely the 
normal probability curve. In this species, though the size-scheme 
is very clearly revealed by quantitative study, the differences 
among the zooids are so small relatively to the whole animal as 
to be unrecognizable to the eye. Consideration of such environ- 
mental and physiological influences as might conceivably bring 
about the observed gradation among the zooids led to negative 
conclusions, leaving no other recognizable explanation of the 
phenomenon than that growth itself proceeds in that way. 
The paper by Ritter and Bailey (1908) records a preliminary 
attack on the same general problem in quite a different and more 
fundamental quarter, but a quarter where the difficulties of 
manipulation are much greater; that is, in the realm of cell 
division as this manifests itself in the developing egg. The 
method of quantitative determination was that of weighing. So 
far as the problem itself is concerned, the meager results were 
