21,1 ° Shaw: Merrillosphaera 111 
Apparently an absence of female coenobia in advanced stages 
of development in Powers’s material was a cause contributing 
to the error in hig case, and a similar condition of the material 
used by Klein may have contributed to the same error in his 
work. Nevertheless, the simplest hypothesis for us to adopt 
for the purposes of the present discussion is that the large re- 
productive cells present in several of Klein’s coenobia were 
in some cases gonidia, and in others the nearly ripe and the 
recently fertilized eggs that he called them. 
The form of the coenobium in four of Klein’s figures is nearly 
spherical, and in the other four ‘decidedly ovoid. The dimen- 
sions ° were given for six of the coenobia. Assuming that all 
of the drawings are on the same scale” I have measured those 
for which he did not give figures and here give the dimensions 
of all the coenobia in the order of the figure numbers: 290 by 
325 p», 345 p, 270 p, 225 by 260 », 320 by 380 p, 290 pz, 
300 », and 320 by 425 ». Of the spherical coenobia, two are 
practically isodiametric and two are very slightly elongated. It 
is the polar axis, if any, that is the longer. 
The somatic protoplasts were represented as round in surface 
view and ovoid in median optical section views of the coenobia. 
It was stated that the cilia were not visible under the magnifi- 
cation that was used for making the drawings. Measurements 
of the vegetative cells, by which I understand Klein to mean 
protoplasts, were given for all of the coenobia as follows, in 
the sequence of the figure numbers: 7.3 to 7.5 p, 6.4 to 7.3 p, 
6.5 u, 7.3. to 8.9: 7 to TH p73. to 8 pm 64 to 7.3 4, 
and 6.4 p». 
The somatic protoplasts are farther apart at the anterior pole 
and the spacing is graded from one pole to the other in the 
drawings of all but one or two of the smaller coenobia. 
The number of somatic cells present was stated by Klein for 
all of the mother coenobia except that of his fig. 4. I have 
assumed that these numbers were obtained by a method which 
Klein stated in his first paper (‘89A, p. 146). The formula 
used in that procedure made no allowance for the fact that the 
areas occupied by the individual cells are * proportional to the 
areas of hexagons having diameters equal to the intercellular 
*The figure 260 given for the shorter dimension of the first coenobium 
is obviously a misprint for 290. 
"The scale is evidently about 152 diameters, and not 120 as stated in 
the explanation of the plate. 
* As was pointed out by Janet (12, p. 28). 
