Page:Proceedings of the Royal Society of London Vol 60.djvu/377

350 (1) To test by actual experiment Duval’s theory of the formation of the primitive streak.

(2) To try and determine experimentally whether the whole or only part of the actual embryo is developed by the activity of the primitive streak. And further, if only a part, to determine its limits.

With regard to the first question it may be remarked that Duval’s account is generally accepted, although perhaps greater stress is laid upon it by foreign and American writers than by embryologists in this country.

According to Duval’s account, there is in the freshly laid and unincubated egg a groove which separates the blastoderm from the yolk. The groove, he says, is broader and more conspicuous at the posterior margin than at any other point. This he compares to the anus of Rusconi or blastopore of the segmenting frog’s egg.

During the first few hours of incubation the edge of the blastoderm is said to advance'over the yolk at every point except at this most posterior margin bounding the groove, which he regards as equivalent to the frog’s blastopore. At this spot there is no advance. The portions of the edge of the blastoderm adjoining this part swing round to meet each other in the middle line, and eventually fuse and form what Duval calls the “ plaque axiale.”

This structure is in reality the primitive streak, and, according to Duval, it becomes visible as such during about the tenth to fifteenth hours of incubation by reason of the subsequent hollowing out of the subjacent yolk by the extension backwards of the sub-germinal cavity.

Such a mode of growth would be very extraordinary and interesting if true, and would be very acceptable to those who believe that the growth in length of the Vertebrate embryo is caused by a concrescence of two at first separated germinal rims.

Naturally this account of the formation of the primitive streak as given by Duval is frequently quoted by the many adherents to the concrescence theory.

During the last few years experimental methods have been introduced much more freely into investigations of animal development. Foremost amongst the workers upon these lines is Dr. Wilhelm Roux, who experimented by destroying certain cells of the segmenting eggs of frogs, and noting the result after some days of development. He has been followed in similar work by Morgan and Ume Tsuda and others.

The eggs of frogs have been the object of experiment of a different