74 THE AMERICAN NATURALIST. (Vor. XXXIX. 
ogeny, which is characterized by a reduced number of chromo- 
somes, and by the presence of the active sperm-centrosome. In 
stimulating unfertilized eggs by chemical or physical means to 
develop, we get what is called artificial parthenogenesis. As I 
have elsewhere (Petrunkewitsch, : 04) tried to show, this develop- 
ment is due not to a new formation of centrosomes,! but to the 
stimulation to new life of the egg-centrosome, which is other- 
wise destined to disappear. According as we apply the stimulus 
before or after the second maturation division, we obtain differ- 
ent results ; if before, we get development with the normal num- 
ber of chromosomes, if after, then with the reduced number. 
Now, it is interesting that both parthenogenetic development 
with the reduced number of chromosomes and merogenic devel- 
opment show abnormalities when compared with the develop- 
ment of fertilized eggs. These abnormalities increase in inverse 
proportion to the number of chromosomes left in the egg, an 
indication of which is given in the experiments of Stevens, 
which consisted in a dissection of the fertilized egg in the 
amphiaster stage into two parts containing unequal numbers 
of chromosomes. All this goes to show that neither merogeny 
nor artificial parthenogenesis with a reduced number of chromo- 
somes can be regarded as equivalent to natural parthenogenesis. 
I call, therefore, these two forms, artificial, pathological, uni- 
parental development. On the other hand, artificial partheno- 
genesis with the normal number of chromosomes leads to appar- 
ently normal development. I therefore call it artificial, true 
parthenogenesis. 
If we could cause an egg to develop by applying a pera en 
before the first maturation division, the gap between partheno- 
genesis and budding would be filled. Again, if we could. bring 
about artificial parthenogenetic development through several 
successive generations by stimulating the eggs at the three 
different moments of their maturation, we might obtain valuable 
‘If there are such de novo formations of centrosomes, as seems to be again 
emphasized by the research of Yatsu (:04), it is nevertheless probable that in 
counteracts possible teratological action on the part of the artificial centrosomes. 
