21,1 Shaw: Merrillosphaera 109 
the periphery of a transverse plane considerably anterior to 
the equator of the coenobium, the four others lie in the pe- 
riphery of a transverse plane well toward the posterior pole and 
alternating with the members of the anterior quartet. When 
ten are present, two are usually uniformly smaller and placed 
between the second quartet and the posterior pole on opposite 
sides of the polar axis in positions more or less nearly opposite 
to members of the anterior quartet. When twelve are present 
there are two quartets of larger gonidia and a posterior quartet 
of smaller gonidia. Deviations in the number of gonidia occur 
by the omission of one or more members of one or more of 
the quartets or pairs. In such cases the remaining members 
retain their typical places. 
The segmentation of the gonidia was not described, and the 
radial flattening of the gonidia which precedes segmentation 
was not recognized as such but was noted as a character of 
the “ova,” that is, the gonidia. The account of the embryonic 
metamorphosis is so involved with an account of an invagination 
and inversion process that it is impossible to decide what of it 
is pathological and what of it, if any, is misinterpretation of 
stages of normal development. Instances of invagination in 
young but completely closed coenobia were sought by Powers 
(08, p. 160), but not found. 
The male coenobia were called dwarf male colonies. Powers 
records having observed one hundred thirty-one parental coe- 
nobia bearing recognizable male progeny. The number of 
males in a parent varied from one to ten. The progeny were 
exclusively male in only one instance, the one shown in his 
fig. 57. The unborn male coenobia are about equal in size to 
their unborn asexual sisters, about 210 » in diameter. They 
are dwarf only in the sense of having no postnatal growth. It 
was stated that the sperms were ripened before the male coe- 
nobia were liberated. The number of somatic cells is small 
in proportion to the number of androgonidia. The sperms are 
formed in tabloid bundles of sixty-four and one hundred twenty- 
eight. The two sizes occur side by side. The sperm bundles 
are slightly concave, presumably on the posterior, nonciliated 
Side. The spermatozoids are slender and have terminal cilia. 
So far as it relates to female coenobia and to real oogonia 
Powers’s account is almost blank. This fact may have a signifi- 
cant bearing on the consideration of Klein’s account of the 
material collected by Migula, which is to be taken up in the 
next section of this paper. In Powers’s inventory of the contents 
