226 BULLETIN 114, UNITED STATES NATIONAL MUSEUM. 



Scale formulae of multidncta. 



Formula. 



El Dorado 

 to Kern 

 County. 



Los Angeles 

 and vicin- 

 ity. 



Vicinity of 

 San Fran- 

 cisco. 



Other locali- 

 ties. 



Total. 



Male. 



Fe- 

 male. 



Male. 



Fe- 

 male. 



Male. 



Fe- 

 male. 



Male. JTate. 



Male. 



Fe- 

 male. 



23-25-23-21-19 

 23-21-19 

 21-23-21-19 

 21-19 

 23-21-19-17 

 21-23-21-19-17 

 21-19-17 

 19-21-19-17 



Total 





2 

 1 

 3 















2 



1 



12 



3 



"3" 

















1 



4 

 1 



1 



1 



"i" 



1 



4 



2 







2 

 1 

 3 

 5 

 11 

 1 



2 

 5 

 1 





















1 



3 



1 



3 





4 

 1 



1 















8 



7 



4 



6 



4 



1 



7 



7 



23 



21 





Ventrals and annuli of multidncta. 



Localities. 



Num- 

 ber of 

 speci- 

 mens. 



Ventrals. 



White annuU. 



Extremes. 



Aver- 

 age. 



Extremes. 



Aver- 

 age. 



El Dorado County to Kern County 



Los Angeles County 



15 



16 



5 



202-222 

 207-218 

 205-210 



215 

 213 



208 



37-57 

 44-51 

 23-40 



46 

 48 

 35 



Vicinity of San Francisco Bay 





Affinities. — The affinities of this form are much in doubt. It was 

 originally described as a subspecies of getulus. More recently 

 Stejneger (1902, 153) rated it as a subspecies of pyrrliomelaena. Tliis 

 is undoubtedly much nearer to its true relation. In favor of this 

 it may be noted that (1) the two forms present a striking superficial 

 resemblance to each other; (2) they occupy adjacent regions; (3) 

 a considerable proportion of the specimens of each possess the scale 

 formula 23-21-19-17, a formula very rare in other forms of the 

 genus. It is generally accepted that two forms to be rated as sub- 

 species must intergrade in the region of the common boundary of 

 their ranges. Multidncta is certainly not now known to intergrade 

 with pyrrhomelaena, and the differences between them are so great 

 as to render it unlikely that they do intergrade. Pyrrhomelaena is 

 much further removed in structural characters from the mean of the 

 group than multidncta, so that we could hardly derive the latter from 

 it; to derive the former from multidncta we must assume the west 

 coast as the center of origin of this group, but, even assuming this, 

 we have far too much specialization in pyrrhomelaena for the short 

 geographical distance that it is removed from its ancestor, and further- 

 more we have great differentiation in structural features with but 



