324 



H. H. NEWMAN 



of some other species. It will be seen, then, that the material 

 is somewhat complex and difficult to control by means of exact 

 experimental manipulations. We are therefore forced to make 

 use of certain natural experiments which fortunately have come 

 to hand and which appear to furnish the data needed for an 

 analysis of the causal factors involved in twinning. 



4 ^6 



(Figs. 1 to 28 are from parthenogenetic cultures) 



Fig. 1 Normal blastula, showing distinct polarity, small cells at apical end, 

 large cells at basal. 



Fig. 2 Half-sized dwarf blastula with indefinite polarity. 



Fig. 3 Fourth-sized dwarf blastula with little or no polarity. 



Fig. 4 Normal early gastrula. 



Fig. 5 Rare case of typical half-sized gastrula, resulting from half-blastula 

 that had acquired a definite unipolar axis. 



Fig. 6 Rare case of quarter-sized gastrula from a polarized quarter-sized 

 blastula. 



The first evidences of twinning that came to the writer's 

 attention appeared in certain hybrid strains in which the eggs 

 of Patiria had been fertilized by the sperm of another asteroid, 

 Pisaster ochraceus. In several cultures of these cross-bred strains 

 there were noticed, on the second day, considerable numbers 

 of very small blastulae and gastrulae, some about one-half and 

 others one-fourth as large as normal larvae at the same stages. 

 A good idea of the size and appearance of some of the more 



