52 PROFESSOR VAN BENEDEN. 



The fundamental idea of this theory is that a part of the tissues 

 of the embryo are not derived from the egg-cell but proceed from 

 maternal cells, which immigrate into the vitellusand are destined 

 to give rise to connective and vascular tissues. The embryo on 

 this theory would be related to the maternal organism by a 

 double connection ; through its archiblastic tissues derived 

 from the egg-cell it would be descended from parental archi- 

 blastic tissues ; through its parablastic tissues it would be 

 derived directly from the mothers connective and vascular 

 tissues. Not only has this theory not been demonstrated by 

 Mr. His, but the idea appears to me to be in formal contra- 

 diction to the facts which are the outcome of comparative 

 embryology. But it is equally erroneously that Kathke (1-i), 

 von Baer^ (15), Strieker (16), lieineck (17), AVeil (18), and 

 CEUacher have maintained that the various germ-layers are 

 formed in the Osseous Fishes by a process of delamhiation at the 

 expense of that part of the egg alone which takes part in the 

 cleavage process. Whilst I cannot admit the view adopted by 

 Gotteand Haeckel — according to which the internal layer is formed 

 by invagination of the margin of the blastodisc, and is accordingly 

 derived like the external layer from the segmented disc — it by no 

 means follows that I should deny that the phenomenon of invagi- 

 nation does take place in the Teleosteans. The following is 

 the mode in which it seems to me necessary to interpret the phe- 

 nomena which cliaracterise the commencement of development. 

 Directly after fcecundation the egg of the Osseous Fish divides 

 into two very unequal cells, very dissimilar, differing in consti- 

 tution and significance ; the one is the germ which segments and 

 from which the blastodisc is derived ; the other is formed by the 

 deutoplasmic globe — clothed, at least partially, by a thin layer of 

 protoplasm forming ''the intermediate layer." This cell is the 

 origin of the endodermic layer of the future embryo ; it has a 

 constitution analogous to that of a fat-cell. This never pro- 

 ceeds to segmentation, but there appears in it at the conclusion 

 of the segmentation of the other cell (the gt-rm) a great lot of 

 cells, which take their rise by " free-cell formation.'^ Hence 

 results the apparition of a cellular layer subjacent to the blas- 

 todisc. The latter is the homologue of the ectoderm of other 

 Vertebrates, the former is the homologue of the endoderm. It is 

 ascertained that the blastodisc extends its area little by little and 

 tends to cover in by "epiboly" — the deutoplasmic globe. The 

 observations of Van liainbeke have rendered it very j)robable 

 that the intermediate layer makes a simultaneous jjrogressive 

 movement. The upshot of this is that the develo])ment of the 

 Teleosteans commences by the total cleavage of the vitellus into 

 two cells ; one of the two continues to segment whilst the other 



