February 4, 1898.] 



SCIENCE. 



157 



diftereut stages. The more complete the 

 series the fuller will be our knowledge. 



Now Hertwig maintains that this knowl- 

 edge of the successive stages of develop- 

 ment is itself causal knowledge bej'ond 

 which we can not hope to go. He holds 

 that the egg is the cause of the blastula and 

 the blastula the cause of the gastrula, etc. 

 Hertwig pretends to be completely satis- 

 fied with knowledge of this sort. 



Comparison is not perhaps always just, 

 yet Hertwig's position is the same, I think, 

 as though a physicist were to say that if we 

 knew the path of the moon around the 

 earth we should know everything that we 

 could hope to know, or if the astronomer 

 claimed that the position of the moon at 

 one moment is the cause of its next position. 



Roux, on the other hand, maintains that 

 in order to understand the successive stages 

 of development we must know how the one 

 transforms itself into the other, how the 

 blastula invaginates to form the gastrula, 

 how the medullary plate of the vertebrate 

 embrj'o rolls in to form a tube. The move- 

 ments, then, of the parts of the embryo are 

 to be studied. But even a knowledge of 

 the movements of cells and groups of cells 

 would not be causal knowledge, although 

 it might, perhaps, be called the mechan- 

 ics of the embryo. What makes the 

 endoderm turn in ? What induces the me- 

 dullary plate to roll up into a tube? What, 

 in brief, are the forces at work ? 



A few illustrations of the kind of 

 work that embryology has already accom- 

 plished may bring before us more clearly 

 the problems of to-daJ^ 



Pflilger 's experiments on the effect of grav- 

 ity on the segmentation of the egg naturally 

 suggest themselves first. When the egg of 

 the frog is inverted, with its dark hem- 

 isphere turned down, the cleavage planes 

 appear, not in their normal position, but in 

 respect to the direction of the force of grav- 

 ity. At first Pfliiger seemed to think that 



there is some causal relation between the 

 force of gravity and the forces that direct 

 the cleavage of the egg. 



Roux showed, however, that a centrifu- 

 gal force could replace the force of grav- 

 ity, and, moreover, that if the experiment 

 were so arranged that the centrifugal force 

 just overcame the force of gravity then the 

 egg segmented normally in whatever posi- 

 tion it was placed. 



Finally Born showed that a rotation of 

 the contents of the inverted egg occurred 

 so that the lighter parts rose to the highest 

 points. It is obvious, therefore, that grav- 

 ity only indirectly affects the egg by bring- 

 ing about a rearrangement of its contents. 



This series of experiments is instructive, 

 I think, in that it illustrates how one ex- 

 periment leads to another, and how our 

 knowledge of the forces at work is ad- 

 vanced with each well-planned experiment. 

 We do not know, to be sure, why the egg 

 segments, but we have found out something 

 definite about the action of gravity on the 

 egg. 



Let us now consider another series of ex- 

 periments : Roux found that by preventing 

 the development of one of the first two 

 blastomeres of the frog's egg the other un- 

 injured blastomere developed into a half- 

 embryo. Naturally enough, he drew the 

 conclusions that the first two cells are self- 

 differentiating, and that the development is, 

 at least in part, a mosaic work. The con- 

 clusion was, I believe, not justifiable at the 

 time, because in the experiment the injured 

 half of the egg remained in contact with 

 the developing half. 



Later experiments on other forms — the 

 Sea-urchin's egg, for example, where the 

 blastomeres can be completely separated — 

 gave other results. A whole embryo de- 

 veloped from the isolated parts. Roux's 

 conclusions were said to be overthrown. 

 Then came an unexpected result. The 

 blastomeres of the ctenophor may be com- 



