1026 MR. G. A. BOULENGER ON 



The specimen in Dr. Spurrell's collection measures 55niiti,'from 

 snout to vent. It was shining jet-black in life, with the belly 

 grey. The thick skin of the ventral region forms a strong fold 

 on each side, which extends across the pubic region. 



16. Ran A palmipes Sjaix. 



A single female specimen, measuring 95 mm. from snout 

 to vent. 



Tympanum three-fourths the diameter of the eye. First 

 finger extending well beyond second. Skin of back finely 

 granulate ; upper surface of tibia with longitudinal ridges 

 formed by confluent granules. Tibio-tarsal articulation reaching 

 half-way between the eye and the tip of the snout. Back pale 

 olive-grey (appears to have been green in life), without spots ; 

 belly uniform white. 



I have examined smaller specimens (40 to 70 mm.) from the 

 same locality (Condoto) from Mr. M. G. Palmer's collection. In 

 the smallest the tympanum measures barely two-thii-ds the 

 diameter of the eye, the upper parts are bla.ckish brown and 

 the lower parts profusely spotted with black. 



17. Phyllobates pratti Blgr, 



A single specimen, 15 mm. long from snout to vent. 



This species was described in 1899 (Ann. & Mag. N. H. [7] iii. 

 p. 274, pi. xi. fig. 3) from two specimens obtained by Mr. A. E- 

 Pratt at Santa Ines, north of Medellin, at an altitude of 3800 feet. 

 Other specimens have since been found in the Choco, at Novita, 

 altitude 150 feet, by Mr.M.G. Palmer. In Dr. Spurrell's specimen, 

 the white line in front of the thigh is produced forward to the 

 upper eyelid, dividing obliquely the black lateral band. 



18. Dendrobates tinctorius Schneid. (PI. CIY. figs. 2-6.) 



The extraordinary colour variations to which this species is 

 subjected are only paralleled by those of the African frogs of the 

 genus Rafijia, and afi'ord the same diti&culties to the systematist. 

 Some of these variations are local, others are merely individual ; 

 great is the perplexity in having to decide which deserve to be 

 recognised by a special varietal name. In the collection made by 

 Di'. Spurrell the specimens fall under two principal categories which 

 I will designate as vars. cocUei and chocoensis respectively. It is 

 not an easy matter to fix the typical form of D. tinctorius ; the 

 following remai-ks will not be superfluous, t(5 justify the course 

 followed by me. 



Schneider (Hist. Amph. i. p. 175, 1799) is the author of the 

 name tinctorkis, but the frog itself was unknown to him. He 

 refers us to Lacepede (Quadr. Ovip. p. 566, pi. xxxix., 1788), 

 who regards the " Raine a tapirer" as a variety of Ilyla rubra 

 Laurenti, and describes it as reddish with two irregular yellowish 

 white longitudii:ial bands ; the figure is hopeless for the purpose 



