298 BULLETIN 58, UNITED STATES NATIONAL MUSEUM. 



form a heart-shaped figure; 3 pairs of chin-shields, posterior shortest; 

 23 rows of small, nonimbricate, convex and keeled scales without 

 apical pits, the soft skin mostly visible between the scales; 165 ven- 

 trals; anal entire; 49 undivided subcaudals. Color (in alcohol) above 

 uniform '' walnut-brown," more rufous toward the middle of back, 

 with a ver}^ distinct black longitudinal stripe along the median line 

 involving the median scale row and the adjacent half of the scale on 

 either side; underside paler, more cinnamon colored, each scute with a 

 narrow light edge and a darker submarginal line; a median dusky 

 stripe on the underside of the tail. 



Dimensions. 



mm. 



Total length 412 



Snout to vent 347 



Vent to tip of tail 65 



The male seems to have a relatively longer tail with more subcaudals. 

 Thus the specimen in the Science College Museum, Tokyo, measures 

 410 mm. in total length, of which the tail is 90 mm. and contains 61 

 subcaudals. 



The young is darker above in stronger contrast to the underside 

 and seems to lack the dark median stripe on the underside of the tail. 



Variation. — Only a few specimens are in the museums, but these 

 show considerable variability. The six of which the individual scale 

 formulas are recorded, agree in having 6 supralabials; no preoculars 

 and no postoculars (except occasionally a trace) ; 2 + 2 elongated tem- 

 porals; the scale rows vary between 21 and 23, six being known with 

 23 and only one, the type, with 21; ventrals between 146 and 166; 

 subcaudals between 45 and 62 ; all have an undivided anal and undi- 

 vided subcaudals. The specimen described above seems to be unusual 

 in having the loreal separated from the posterior nasal by the pre- 

 frontal touching the second supralabial. The specimen figured (fig. 

 259) is also abnormal in possessing only two pairs of chin-shields. The 

 Tokyo museum specimen shows a curious apd rare anomaly in having 

 the frontal divided longitudinally by a suture, and it is interesting to 

 note that one of Sauvage's types of his A. hraconnieri, from China, 

 also has the frontal divided (though he does not say whether longi- 

 tudinally or transversely) ; this specimen also has one of the supraoc- 

 ulars divided. 



Hahitat. — Under the supposition that ^4. hraconnieri, which is 

 recorded from eastern Kiangsi and from Ichang, in southern China, 

 represents a distinct form, the present species is restricted to Japan. 



Thus far only ten specimens are known, seven of which are from 

 definite localities, which throw some light on the distribution of the 

 species. Of these three are from Kiusiu, viz, the two which Mr. N. 

 Takachiho collected at Hikosan, province of Buzen, near the northern 



