EGG SECRETIONS OF ARBACIA AND ASTERIAS 469 



secretion obtained August 11, 1914, tested to only 50-units 

 strength while that obtained by exactly the same method early 

 in the season averaged 1600 to 3200 units and frequently ran 

 higher. 



b. The secretion, as stated on page 45, activates the sperm, at- 

 tracts, reversibly agglutinates, and may finally paralyze them. 



c. The secretion has equally striking effects on the egg. Glaser 

 ('14 c) observed that Arenicola larvae treated with egg secretion 

 lose pigment, indicating an increase in permeability. Taking 

 his cue from this, and from the fact that parthenogenetic agents 

 as a class increase permeability, he found that when the secretion 

 derived from either Asterias or Arbacia was added to ripe eggs of 

 the same species, and allowed to act on them for two hours, the 

 eggs began to segment. To this process he gave the name 'auto- 

 parthenogenesis.' Following the method of Loeb ('13, p. 70), 

 Glaser found that after-treatment with hypertonic sea-water 

 (50 cc. sea-water and 8 cc. 2.5 m. NaCl) increased the percentage 

 of developing eggs. 



Glaser's work on autoparthenogenesis was easily confirmed. 

 The 'dry' ripe eggs of Asterias were placed in watch-glasses with 

 at least two volumes of standard fertilizin for one and one-half to 

 three hours. They were then rinsed two or three times with sea- 

 water, and treated with hypertonic sea-water (50 cc. sea-water 

 and 8 cc. 2.5 m. NaCl) for a suitable length of time. After rinsing 

 with sea-water, they were left to develop. It is better to transfer 

 them to finger-bowls with a quantity of sea-water (table 3). 

 The optimum time for the exposure of the eggs to fertilizin and 

 for the hypertonic after-treatment seemed to vary with the tem- 

 perature and with the batch of eggs. Occasionally the entire lot 

 of eggs would give normal cleavage and normal gastrulae. More 

 often, however, cleavage of both nucleus and cytoplasm was irreg- 

 ular, as described by Wilson ('01) for parthenogenetic eggs. 

 If, when the cytoplasm first divided, the nucleus went entirely 

 into one half, the other half would disintegrate and only the 

 nucleated part would develop into a ciliated blastula. Conse- 

 quently, it was a frequent occurrence to find a small Asterias 

 blastula swimming around among cytoplasmic fragments within 



