248 T. H. Buruenp, 
only) which ought to be segmental in position, yet the section con- 
taining it passes between the ninth and tenth somites. Rasu 
himself admitted that the pronephric “Anlage” may be quite prominent 
intersegmentally. It seems that a groove-like origin of the rudiment 
is much more in accordance with the earliest appearance of the 
Selachian kidney, and that it is not possible to distinguish nephro- 
stomes until a later period. A series of successive sections of the 
pronephros in all early stages shows that the structure is un- 
doubtedly unsegmented and continuous. 
Such a view certainly appears more satisfactory than that of 
Rast, to accord with which such a statement as the following is 
made — “The number of tubules exactly corresponds with the number 
of somites over which the pronephros extends. The occurrence of 
more than the corresponding number of tubules as regard the somites, 
is explained as due to degeneration setting in early”. 
The so-called Sammelgang (Sammelrohr). 
Rast alone describes this, van WIJHE did not apparently notice 
it. A careful examination of my sections has not revealed any 
structure such as-is shown in Rast’s figs. 3, 6, 11 etc. on tab. 13. 
The incipient pronephric duct (in the pronephric region) is quite 
indistinguishable from the remainder of the “Anlage” in the earlier 
stages. On this matter RABz writes — “In der That verbindet dieser 
Strang (Vornierenstrang) die Kuppen der einzelnen Hervorwülbungen 
der lateralen Mesodermlamelle, aus denen der ganze Vornieren- 
wulst zusammengesetzt ist” — though somewhat earlier he writes — 
“dabei geht er (Strang) mit diesem (Wulst) mehrfache Ver- 
bindungen ein”. — 
Why are these connections numerous, instead of being metameric ? 
Van WiIJHE, on the other hand, says that the Sammelgang in 
Pristiurus is formed from the first along with the tubules out of 
mesoderm. 
Both van WusHeE and Rast tried to show that the pronephric tubules 
are directed towards one another distally, thus suggesting a fusion of 
their outer ends: but even in this no uniformity in the behaviour of the 
tubules could be detected. RaBz states that the anterior ones are 
usually directed caudalwards, the middle ones lateralwards and the 
posterior ones cranialwards; but the posterior tubules may be 
directed lateralwards (RaBz) and the anterior ones cranialwards 
(van WIJHE). If it be true that the tubules are undifferentiated at 
