REGENERATION OF LOST PARTS IN ANIMALS. 47 
course of subsequent growth. The direction of the cut 
counted for something. If a posterior portion of an earth- 
worm be cut off obliquely, there is on the part of the 
anterior portion a sort of sloughing which effects a straight 
plane for regrowth of the missing hind piece. But if the 
worm be decapitated obliquely, an anterior bud grows out 
at right angles to the surface of the wound, and is 
gradually brought into its proper position as growth goes 
on, unlike what occurs in TJwbularia. The variety in 
different cases seems to suggest that the result is, as it were, 
a compromise between the inherent growth-tendencies of the 
organism and the environmental stimuli operative in each case. 
I have left to the end any notice of one of the most 
puzzling facts which the recent study of regeneration has 
brought to light—the fact that the regenerative growth is 
sometimes very different from that which occurs in 
embryonic development. Two examples must suffice. 
When the anterior end of a Naiad is cut, an ectodermic cap 
is formed, according to Hepke, over the wound; in the 
concave interior of this cap all the organs to be replaced 
gradually appear; muscles, which are normally mesodermic, 
are formed by cells migrating from the ectoderm, and a gut 
apparently ectodermic grows back to meet the severed 
endodermic gut (Zool. Anzeig., xix. (1896), pp. 513-6). 
In the Annelid, Ophryotrocha puerilis, as described by Rievel 
(Zeitschr. wiss. Zool., |xii. (1896), pp. 289-234. 3 pls. 1 
fig.), the very front of the new gut formed in regenerating 
the results of decapitation is made by a forward growth of the 
old endodermic gut, whereas normally the fore-gut should, 
of course, be ectodermic. According to Michel, the caudal 
bud which regenerates a curtailed Nephthys is wholly 
ectodermic (Comptes Rendus, Acad. Sci. Paris, exxiii. 
(1896), pp. 1015—7, 1080-2). In Tubifex, the anteriorly 
regenerated fore-gut is endodermic all but a small, most 
anterior piece (Haase). It is plain that regenerative growth 
does not necessarily follow the path of embryonic develop- 
ment, but the same sort of difficulty arises in connection 
with the buds of Tunicates and Polyzoa, and is perhaps a 
hint that we tend in embryology to make too much of the 
distinctiveness of the germinal layers. 
