804 MR. R. I. POCOCK ON THE 



was adopted by Mivart (P. Z. S. 1885, pp. 393-395); and with 

 the addition of Pcecilogale and Lyncodon to the Mustelinee was 

 repeated by Flower and Lydekker in 1891 (' Mammalia,' pp. 567- 

 591). Tt was also accepted by Sedgwick (' Students' Text-Book 

 of Zoology,' ii. p. 624, 1905). 



In 1902 de Winton ('Zoology of Egypt, Mammalia,' p. 245), 

 trusting apparently to the structure of the teeth, modified 

 Flower's system by transferring MejJiliitis, Conepatus, Mellivora, 

 and Ictonyx from the Melinee to the Miistelinse. Max Weber in 

 1904 (Die Saug. pp. 536-538) followed de Winton with respect 

 to the allocation of Mellivora, but otherwise agreed with Flower. 

 And in the same year Trouessart (Cat. Mamm., Suppl. pp. 187- 

 215) also followed Flower, but put Pcecilogale into the Melinse. 

 He further subdivided Mustelinee into two tribes : — Guloneee for 

 Gulo, Galictis [^■=Tayra-\-Grison~], and Lyncodon, and Mustelese 

 for Mustela [=jlfarfes]and Putorius \^=^ Mustelci], with subgenera 

 Luireola and Arctogale \_-=.Mustelci\. 



In 1912 Miller (Cat. Mamm. Western Europe, pp. 340-341) 

 divided the Mustelida^ into four subfamilies : — (1) Melinse con- 

 taining " about a dozen genera," of which only Meles is quoted and 

 dealt with ; (2) Lutrinee containing " four genera," Lutra alone 

 being discussed ; (3) Guloninse containing Gido and probably 

 Mellivora ; (4) Mustelinse restricted to the genera Martes, Mustela 

 (with its subgenera Putorms and Lutreola and Mustela), and 

 Vormela *. 



Although Miller, like de Winton, seems to have relied mainly 

 on the teeth, it seems clear that a large number of the genera 

 placed by de Winton in the Mustelin?e were left by Miller in the 

 Melinse, where Flower placed them. Another point to be noticed 

 in connection with Miller's classification is his severance of Gulo 

 from the Mustelinfe as the type of a special subfamily f, with the 

 guess that Mellivora is a related genus. 



Setting aside for the moment the classifications of Gray and 

 Gill, two conclusions come out from this brief review : — (1) There 

 is complete agreement amongst recent writers with regard to the 

 status of the Lutrinse ; (2) Although Flower's subdivisions have 

 held the field until the present day, there is a great variety of 

 opinion as to the constitution of the Melinse and Mustelinse. It 

 is admitted by all that Meles, Arctonyx, Taxidea, Mydaus, and 

 IJelictis are Melinse ; and that Martes, Mustela (with related 

 generic or subgener-ic forms, such as Vormela, Putorius, Lutreola), 

 as well as Tayra, Grison, and Lyncodon are Mustelinse. But 

 Mephitis, Spilogale, Conepatus, Mellivora, Ictonyx, and Pcecilogale 



* In 1911 Satunin (Mitt. Kauk-Museums, v. p. 267) defined the genera Vormela 

 and Putorius, admitting the following subgenera of the latter, Ptitorius (s.s.), 

 Lutreola, Ictis, and KolonoJcus. The genus Putorius is equivalent to Mustela as 

 understood by Miller, Ictis ( = subgenus Mustela) being employed for the stoats 

 and weasels. Miller was perhaps not acquainted with this paper when he wrote his 

 ' Catalogue of the Mammals of Western Europe.' 



f This was originally done by Gray in 1825 (Ann. Phil. xxvi. p, 339); but in his 

 latest classification he abandoned the group named tjruloninge. 



