14 PROCEEDINGS OF THE NATIONAL. MUSEUM vol.81 



distant approach to the condition found in Asiatic specimens (pi. 1, 

 fig. 1). In one specimen (pi. 5, fig. 1), the lateroposterior narrow 

 black edging of the orange spot is abruptly expanded at the end, 

 the swollen end lying almost entirely within the lower portion of 

 the ocellus as in P. m. hudsoniaivus (pi. 3). 



On the under side of the hind wings the marginal lunules are very 

 variable in size, and in most of the sj)ecimens are larger than in the 

 one figured (pi. 2, fig. 2). In the specimen (pi. 5, fig. 3; pi. 6, fig. 3) 

 with the dark border narrowest on the upper surface (referred above 

 to petersii) the inner margin of the dark border below is similar to 

 that in the type specimens of petersii and of hudsonianus. In the 

 other specimens it varies from the condition seen in the type specimen 

 of petersii to the condition seen in the specimen of aliaska figured 

 (pi. 2, fig. 2), being in most cases about halfway between the two. 



Comparisons. — As is evident from the figures, P. m. aliasha re- 

 sembles P. m. sikkimensis more closely than it does P. m. hudsonianus. 

 As was pointed out by Edwards in 1882, both aliaska and sikkimensis 

 are strongly melanic. Both have the dark border of the hind wings 

 above broad, nearly reaching the cell, but narrow below, where the 

 inner border is formed of widely discontinuous black lines, and both 

 have a broad submarginal band instead of a row of isolated spots 

 on the under side of the fore wings. Specimens from the mountains 

 of western China (pi. 1, figs. 1, 2) are more or less intermediate be- 

 tween the two, and Doctor Holland is probably correct in reporting 

 aliaska from northeastern Asia. 



As it occurs in Tibet and western China, sikkimensis is always 

 readily distinguishable from aliaska by having the orange anal 

 ocellus edged above with a broad black border including a thin blue 

 crescent. In aliaska the blue crescent usually lies partly on the lower 

 half of the rather broad black border and partly on the upper portion 

 of the ocellus itself; but it may lie entirely on the black border, or 

 it may have a poorly defined proximal border, scattered blue scales 

 occurring over the upper half of the ocellus. 



SUMMARY 



In northern North America Papilio machaon is represented 

 in the region between Hudson and James Bays and Lake Superior 

 by a relatively large short-tailed form, P. m. hudsonianus^ much like 

 typical P. m. machaon from northern Europe, which appears to show 

 but little variation. 



In Alaska, Yukon, southwestern Mackenzie, and northwestern 

 British Columbia there is a second form, P. m. aliaska^ most closely 

 related to a group of Asiatic forms of which P. m. sikkimensis may 

 be taken as an example. Judged from the meager information avail- 



