PRODUCTION OF TWINS AND DOUBLE-MONSTERS 327 



Figure 1 1 shows a case in which a nearly normal blastula is formed 

 from one blastomere while the other blastomere has remained 

 uncleaved. The ability of one blastomere to undergo cleavage 

 while the other lacks this ability suggests that the products of 

 the first cleavage are physiologically different. It is well known 

 that the first cleavage in echinoderms divides the egg into the 

 prospective right and left halves of the body. In the cases under 

 discussion one side is evidently endowed with a greater degree 

 of developmental energy than the other, and this may be an 

 early indication of the characteristic unilaterality in development 

 that subsequently expressed itself in the asymmetrical devel- 

 opment of the two sides prior to metamorphosis, and the conse- 

 quent change from bilateral to radial symmetry. This is perhaps 

 a far-fetched conjecture, but other data subsequently brought 

 out serve to lend it considerable support. A good many inter- 

 esting conditions were noted in early cleavage stages. In some 

 cases cleavage was evidently proceeding separately in the case 

 of the derivatives of the first two blastomeres. A very clear case 

 is shown in figure 7, which shows the cleavage of one blastomere 

 in a completed four-cell stage and that of the other blastomere 

 in a two-cell stage. Such an early condition probably results 

 in twin blastulae only slightly different, as in figure 8. Other 

 cases occur, as in figure 9, in which the products of cleavage of 

 one blastomere are in a normal four-cell stage, while those of the 

 other blastomere are irregular and are showing signs of disorgan- 

 ization. Almost every kind of isolation of groups of blasto- 

 meres is to be noted, and it is not uncommon to find minute 

 balls of ciliated cells attached to an unorganized mass of large 

 cells. A great many instances of exogastrulation occur, but 

 these do not concern us in this place. 



c. One of the first evidences of twinning in Patiria was based 

 on the discovery of numerous dwarf larvae, chiefly blastulae, 

 which were only one-half or one-fourth the normal size (figs. 2, 

 3, 5, 6) . At the time of this first discovery there was no evidence 

 to confirm the suspicion that these larvae had originated by a 

 process of blastotomy. Subsequently, however, the confirmatory 

 evidence appeared, and now we feel safe in concluding that 



