1891] Zoology. 741 
extreme tip of each branch. In each three vein-like structures extend 
to the tips of branches, which I suppose to be the spinal cord, aorta, 
and caudal vein; but, as neither specimen has been dissected, and 
these structures are not very clearly defined, this point must rest in 
doubt. 
The lines which cross near bifurcation appear to do so in each speci- 
men, a dorsal branch of chorda passing into ventral, and a ventral 
branch into the dorsal fin-fold. 
So far as known, all duplications of this kind have been to the right 
and left of a median line; as in Japanese goldfishes, and this vertical 
duplication, with the result of appending to an ordinary tadpole a struc- 
ture closely resembling the forked caudal fin of some fishes, may suggest 
one of the methods of evolution of fishes and frogs from the same 
ancestral vertebrate form. xy 
I should be pleased to correspond with any one interested, and to 
communicate further if other facts can be obtained by a closer exami- 
nation of this exceedingly rare and unusual larval form. 
[Nore.—These cases of abnormal tadpoles described by Mr. Sher- 
wood have seemed to me so remarkable that it appeared very desir- 
able that they should be figured. The method of duplication of the 
tail is precisely the reverse of that observed in the case of the so-called 
‘‘fan-tail’’ races or double-tailed goldfishes from Japan and China. 
What disturbances of ontogenetic processes may have led to the devel- 
opment of this singular form of monstrosity in the tails of tadpoles 
remains to made out. The origin of such irregularities may be 
coupled with actual mutilations, as seems to be indicated in other cases, 
by the experiment of cutting off the tails of tadpoles, as described 
in the Archiv f. mik. Anatomie, 1891 (D. Barfurth on functional adap- 
tation and the regeneration of tissues in the Amphibia). In the 
memoir referred to it was found that the angle, with reference to the 
notochordal axis, at which the tip of the tail of a tadpole was cut off 
determined the direction of the inclination, upwards or downwards, of 
the tip of the tail, which was reproduced. If the tail was cut 
square across or at right angles, there was no departure from the nor- 
mal form of the reproduced tip. If, however, the tip of a tadpole’s 
tail was cut off so that the upper half of the plane of section, or that 
above the notochord, formed an acute angle with the latter (the angle 
` opening forwards), the now newly reproduced tip of the tail would 
have its axis directed upwards. If the lower half of the plane of 
section formed an acute angle with the notochordal axis (the angle 
opening forwards), the tip of the tail which would now be reproduced 
