116 B Canadian Arctic Expedition, 1913-18 



distributed farther north, but the genus, nevertheless, has reached the eastern 

 part of Siberia and Manchuria from where it extends to Alaska, until it reaches 

 the northern and middle regions of this c6ntinent where, from the Atlantic to 

 the Pacific, the genus exhibits its widest distribution. 



It would seem quite natural to consider the wooded belts of the middle 

 parts of this continent to represent a geographical centre of Moneses; possibly 

 the genus migrated from there to eastern Siberia, Manchuria, and the Altai 

 mountains, rather than vice versa. 



Absent from arctic Siberia but recorded from all the other districts enumer- 

 ated on the, table (Table 6), Pyrola minor thus shows the widest distribution of 

 all the members of Pyrolaceae. On this continent the habitat is given as "cold 

 woods, Labrador, White Mountains of New Hampshire, Rocky Mountains from 

 New Mexico and northward to the Barren country from Lat. 64° to the Arctic 

 Islands." Nowhere is it alpine, however, and in the temperate parts of both 

 Worlds it is associated with the same species of Pyrola, and, not infrequently, 

 also with Moneses. 



It seems barely possible to suppose that the species was formerly repre- 

 sented in arctic Siberia, thus having been a member of the circumpolar flora, 

 even if it is known to occur very near Lat. 70° on the west coast of Greenland 

 (Hartz), and on Melville peninsula (Parry). It is, however, a rare plant in the 

 arctic region and, being decidedly a sylvan type, the geographical centre must 

 have been located south of the polar regions, in the wooded belts of the Old 

 World, presumably, since it is much more abundant there than on this continent, 

 and more evenly distributed from north to south. Like Moneses it is very 

 constant in habit, only one variety having been recorded, var. brevis Lge. from 

 Greenland. 



An almost corresponding distribution is shown by P. secunda, with exception 

 of its absence from arctic America and Kamtchatka; in Greenland the' typical 

 plant does not occur, but is replaced by the variety obtusata Turcz.; this variety 

 is also recorded from Terra Tschuktchorum, from Canada, the Atlantic and 

 Central United States. Like the preceding species, P. secunda is a woodland type 

 and is not known to be alpine. Besides the variety obtusata, two others have 

 been described, viz.: pumila Cham, et Schl., and dispersiflora Norm.; of these the 

 former is distributed from Labrador to Alaska; it follows the Rockies south to 

 Colorado, and has been recorded from eastern" Siberia and Altai. The var. 

 dispersiflora Norm, is known only' from arctic Scandinavia. Considering the 

 relatively wide distribution of the typical plant on this continent in the north 

 as well as in the south, extending to California, Colorado, and Maryland, and 

 in view of the development of two characteristic varieties, it appears as if some 

 important centre must have existed on this continent, presumably in the 

 wooded belts of the northern part. With regard to the Siberian distribution of 

 P. secunda, we have seen that this extends from Terra Tschuktchorum, eastern 

 Siberia to Altai, and, furthermore, Maximowicz has recorded it from East Man- 

 churia. But the absence of the species from the boreal regions of Siberia, except 

 in the northeastern corner, makes me believe that the Asiatic element of the 

 species has really originated on the American continent. 



Two important centres of distribution must undoubtedly be attributed to 

 P. chlorantha, as shown on the accompanying table (Table 6). On this continent 

 the distribution extends from Newfoundland, Labrador, and the Maritime Pro- 

 vinces westward to northern British Columbia and Alaska, south to California, 

 Colorado, and-Maryland. In Europe P. chlorantha has reached the arctic zone 

 in Finmark; it is quite generally distributed through the European continent, 

 going as far south as the Alps, the Pyrenees, and Kasan in Russia. But so far, 

 the species has not been observed in Asia, and it is thus the more interesting to 

 notice that an analogous species does exist in eastern Manchuria, namely P. 

 renifolia Maxim. It would be difficult to combine the European distribution 



