424 Illinois ^litate Laboratory of Natural History. 



the same position with respect to the posterior quartet of cells 

 that they do in the 16-cell colony, but are as a rule much 

 larger, the posterolateral pair measuring from 20 to 30 /< in 

 length, while the median one reaches only a length of 1 5 to 

 18 yw. The five tails do not lie in one plane, but share in the 

 spiral of the colony, at times, indeed, exceeding it in the 

 degree of the twisting. These structures are all subject to 

 considerable variations and irregularities of development 

 (Fig. 5}, such as suppression, inequality of members of pairs, 

 differences in size and relative development, in attenuation, 

 and in degree of divergence. These irregularities are often 

 correlated with the loss of cells in the colony due to parasites 

 or other causes. The tails, nevertheless, exhibit such a con- 

 stancy of position and so much of symmetry and regularity 

 of development that they cannot for a moment be considered 

 as ephemeral features of little structural importance. In 

 their position they recall the protuberances noted by me 

 ('98) at the posterior end of Pleodorina illinoisensis. In 

 Pleodorina, however, these structures are apparent only in 

 disintegrating maternal colonies, and it may be that they 

 also indicate the point at which the embryonic cup closes. 

 On the other hand, in Platydorina caudata these tails are 

 present upon the colonies at the time of their escape from the 

 maternal matrix and persist throughout the life of the adult, 

 being permanent structures, characteristic of the species. 

 Within the outer sheath is a homogeneous gelatinous matrix 

 If (Fig. 1, m.) which in Delafield's haematoxylin stains less 

 readily than the sheath. In the living colony no differentia- 

 tion of this matrix is to be seen, but, after staining, a delicate 

 sheath showing deeper color is demonstrated about each of 

 the cells. In most places a considerable space intervenes 

 between this secondary sheath (Fig. 4, s. sh.) and the inclosed 

 cell, so that the sheaths crowd upon each other and appear 

 to divide the field of the matrix into irregular polygonal areas. 

 These areas, as a rule, fill the greater part of the plate, leav- 

 ing unoccupied only a few corners, principally about the 

 second transverse row. The two poles of this swollen second- 

 ary envelope are not of equal size, the inner being somewhat 



