PRODUCTION OF TWINS AND DOUBLE-MONSTERS 325 



nearly normal types of these dwarf larvae may be gotten from 

 figures 2, 3, 5, and 6, as compared with the normal (figs. 1 and 4), 



These larvae were too small to be interpreted as retarded larvae, 

 for they were less in diameter than the unsegmented egg. 

 Neither could they be due to chance admixtures of larvae from 

 another species. It was, therefore, assumed as a working hypoth- 

 esis that they were twin larvae derived by the physiological 

 isolation of the blastomeres of the two-cell and four-cell stages. 

 This assumption was substantiated by a detailed study of clea- 

 vage and early development. During this study there came to 

 light several other types of twins, and these proved more signi- 

 ficant than those first mentioned. As the work extended, it 

 developed that twins of one or more types appeared under three 

 distinct conditions : a) as the result of the spontaneous partheno- 

 genesis of a small percentage of eggs ; b) as the result of crossing 

 Patiria eggs with the sperm of other echinoderm species, and, 

 c) as the result of crowding the normally fertilized eggs of Patiria. 

 It is now proposed to present separately the data derived from 

 each of these types of experiment. 



The illustrations for this paper represent camera drawings of 

 quieted individuals. Only structures that are primarily of in- 

 terest in connection with twinning are shown. 



A. TWINS IN PARTHENOGENETIC CULTURES 



As was brought out in a recent paper (Newman, '20 a), some 

 in practically every lot of Patiria eggs, in which every precaution 

 has been taken to avoid accidental fertilization, undergo sponta- 

 neous parthenogenetic development. The percentage of matu- 

 rated eggs that at least begin cleavage under these conditions 

 varies from in an occasional culture to about 75 in one culture. 

 The usual number ranges from 1 to 10 per cent and is most 

 commonly 2 or 3 per cent. These eggs that are destined to 

 develop parthenogenetically undergo maturation, as may be de- 

 termined by the disappearance of the germinal vesicle within 

 an hour or two after the eggs are shed into sterile sea-water. 

 After maturation nothing further happens for from five and one- 

 half to six hours, following which lapse of time the eggs, without 



