14 G. A. BOULENGER 



The subcaudals are sometimes partly single, as in some true 

 Lycodonts, and I have therefore no hesitation in uniting the 

 genera Boodon and Holuropholis, as I have united Lycodon and 

 Cercaspis. The distinction of the species of Boodon has always 

 been a matter of difficulty. The result of the revision I have 

 recently undertaken is that nine species deserve to be recognized; 

 I have prepared the following synopsis for their identification : 

 I. Subcaudals in two rows (Boodon^ D. & B.). 



A. Praeocular not extending to upper surface of head. 



1 . Scales in 21 or 23 rows ; rostral about once and a half 



as broad as deep. 



a. Frontal as long as parietals. 



1. B. lemniscalus, D. & B. — Abyssinia. 



b. Frontal shorter than parietals. 



2. B. geometricus, Schleg. (= B. seychellensis , Gthr.). — Sey- 



chelles. 



2. Scales in 23 or 25 rows; rostral nearly twice as broad 



as deep. 



3. B. infernalis, Gthr. — East and South Africa. 



B. Prseocular extending to upper surface of head and often 



in contact with the frontal. 



1. Scales in 23 rows. 



a. Subcaudals 47-61. 



4. B. virgatus^ Hallow. (= B. nigrum, Fisch. , B. quadrivir- 



gatum, Hallow., B. ventralis, Gthr.). — West Africa. 



b. Subcaudals 72. 



5. B. guttatuSj Smith (= Alopecion annulifer, D. & B.). — South 



Africa. 



2. Scales in 25 to 31 rows. 



a. Two pairs of chin-shields in contact with each other, 

 a. Parietal shields longer than the distance between the 

 frontal and the end of the snout. 



6. B. lineatus^ D. & B. (= B. capense^ D. & B., B. quadrivittatunij, 



Hallow., B. quadrilineatum, A. Dum., Alopecion varie- 

 gatum^ Bocage, B. biprosocularis ^ Gthr.). — Tropical 

 and South Africa. 



