874 



SCIENCE. 



[N. S. Vol. XXIII. No. 597. 



Notice of the coming botanical symposium 

 to be held from July 2 to 9, 1906, at Mountain 

 Lodge, Little Moose Lake, Old Forge, N. T., 

 was read. 



Mr. H. A. Gleason presented a paper illus- 

 trated by many photographs, on ' Some Phy- 

 togeographical Features of the Prairies.' 



An eastern extension of the great western 

 prairies reaches across Iowa into Illinois and 

 Indiana and portions of the adjoining states. 

 Its flora is characterized by large numbers of 

 western plants, although a majority of the 

 species are of eastern distribution and consti- 

 tute a derived element of the flora. The origin 

 of the prairies has been referred to the char- 

 acter of the soil, the distribution and amount 

 of rainfall, the direction of the prevailing 

 winds, the grazing of bison and to forest fires. 

 Each of these has probably had some influ- 

 ence in accelerating or retarding the invasion 

 of the prairie or forest after the retreat of the 

 continental ice sheet, but the most important 

 factor of all is historical rather than physical 

 in nature. At the close of the glacial period 

 the territory since occupied by prairies was 

 opened first to invasion from the southwest, a 

 region of climatic prairies, and subsequently 

 to invasion from the climatic forests of the 

 southeast. The two floras, on meeting, ad- 

 justed themselves to each other and to the 

 physical factors of the environment, so that 

 ■the forests occupied the bluffs and valleys 

 along the streams, and the prairies the high 

 lands between them. The climate and soil 

 were adapted to the growth of the forest, so 

 that, until extensive cultivation was begun, 

 the prairie was gradually being displaced. 



A comparatively restricted area along the 

 Illinois River is occupied by sand deposits 

 covered with a vegetation essentially similar 

 to that of the sand-hill region of Nebraska, 

 and entirely different from that of the dunes 

 at the head of Lake Michigan. 



After an interesting discussion of Mr. Glea- 

 son's paper, Dr. Eusby exhibited various plants 

 used as food by the Indians. Among these 

 were young shoots of the cat-tail, specimens 

 of bitter-root used by the Indians of the north- 

 west, and Kouse — which consists of several 

 species of Lomatium {L. Caribyi and L. Kous) 



and is an important article of Indian diet. 

 Dr. Rusby also spoke of the use by the Indians 

 of the young buds of the beech tree, which are 

 edible, when cooked, at any date after the first 

 of January. 



Dr. N. L. Britton exhibited fruits of the 

 palm Acrocomia media Cook, recently col- 

 lected by him in Porto Rico,, and remarked 

 on the relationships and distribution of this 

 species, referring to the fine specimen of the 

 plant growing in the palm collections of the 

 garden, brought by Mr. Percy Wilson from 

 that island several years ago. He stated that 

 his observations on this tree showed that the 

 trunk does not invariably bulge above the base 

 as thought by Mr. Cook at the time he de- 

 scribed the species {Bull. Torrey Club, 28: 

 566), a small proportion of the trees being 

 quite colimmar from the base up. He further 

 reported that the Acrocomia of St. Kitts, 

 collected by Mr. Cowell and himseK in 1901 

 is identical with the Porto Rico species, and 

 that it also occurs on the French Antilles, as 

 illustrated by specimens received from Pere 

 Duss. The tree is altogether different from 

 the spindle-shaped Acrocomia fusiformis of 

 Cuba, and seems to be more closely related to 

 the Jamaican A. aculeata. 



0. Stuart Gager, 



Secretary. 



THE PHILOSOPHICAL SOCIETT OP WASHINGTON. 



The 617th. meeting was held on April 7, 

 1906, with President Abbe in the chair. 



Mr. W. D. Lambert presented 'A General- 

 ized Trigonometric Solution of the Cubic' 



In the trigonometric solution of the cubic 

 each case is solved by a device peculiar to 

 itself. If an attempt is made to apply any 

 method to a case for which it is not intended 

 the angles corresponding to the trigonometric 

 functions become complex or imaginary. If, 

 however, we have a means of calculating these 

 angles from tables of trigonometric and hyper- 

 bolic functions the method could be general- 

 ized, and would apply even when the coeffi- 

 cients are complex. Expressions — not new, 

 but not generally known — for the angle cor- 

 responding to a complex sine were deduced. 

 By the use of these, the method heretofore 



