752 JOURNAL, BOMBAY NATURAL HISTORY SOCIETY, Vol. XVI. 



The first note is by Mr. Wasey from Marmagao, describing th? capture 

 of a bull-frog. He notices the eagerness of the shrew to recapture the frog 

 when separated, and the ultimate removal of its dead victim. 



The second note is written by Major Frail, I.M S., and describes the finding 

 of a toad, under the steps of the Residency at Baroda, in the grip of a muskrat. 



The shrew had it between the eyes and was holding on like a bull-terrier. 



The remains of other toads were found in the same place. 



It would appear therefore that occasionally the shrew departs from his diet of 

 insects, and takes to flesh eating. 



W. B. BANNERMAN, Lt.-Col., I.M.S. 

 Plague Research Laboratory, 

 Par el, 7 th August, 1905. 



No. XVII— DOUBLE-HEADED SNAKES. 



With reference to a note on Double Headed Snakes contributed by me to this 

 Journal (Vol. XVI, p. 387), I have just acquired a very interesting paper on 

 this subject which appeared in the Transactions of the Wisconsin Academy of 

 Sciences, Arts and Letters (Vol. XIII, Part II, 1901) by Mr. R. H. Johnson. 

 The writer gives skiagrams of 13 specimens collected from various museums in 

 America, and refers to 17 other instances of this abnormality culled from 

 literature dating as far back as 1640. 



As the magazine in which this appears probably does not come under the 

 notice of the majority of our readers, a few excerpts from this very complete 

 paper will doubtless prove interesting. 



In all the 30 cases the cephalic extremity was reduplicated, and in 3 of these 

 the caudal end was also bifurcate. 



Three examples occurred in individuals of the same brood recorded by Mitchill 

 in 1826, and 2 of these were reduplicated posteriorly. 



The vertebral division, judging from the skiagrams, is always considerably 

 more extensive than is indicated by the cutaneous attachment, and this was the 

 case in the Fyzabad specimen I recorded. 



In 3 of the 13 specimens the skulls were confluent posteriorly, and in the rest 

 the heads and necks were distinct Separate heads and necks are also the rule 

 in the other cases cited, but figures cannot be quoted owing to some of the 

 descriptions being insufficient. In 2 of the 3 examples with confluent skulls, 

 the apposed parietal shields are blended into one, but in the third this point is not 

 manifest. In 3 of the 13 specimens the skulls are malformed, the lower jaw or 

 eyes or both being deficient. 



In 2 of the 13 there is an angular vertebral projection as in the Fyzabad 

 specimen, but the skiagrams show that this projection does not occur at the 

 exact site of the vertebral blending, but at some little distance behind 



The most extensive reduplication of .the 13 shows 67 cervical vertebra) on 

 unc side and 72 on the other, but the specimen recorded by Redi in 1684 is 



