Decembeb 6, 1907] 



SCIENCE 



795 



make thorougli field study ou these interesting 

 water-weeds. 



The study, as far as it has been done now, 

 has given the following suggestions and con- 

 clusions, mostly drawn from the literature on 

 the subject and from herbarium material. 

 There seem to be more than one species, prob- 

 ably six or seven. As far as the material on 

 hand shows, the plant with broad and obtuse 

 leaves, originally described as Elodea cana- 

 densis ^ seems to be hermaphrodite; the others 

 all dioecious, not polygamous. The plant 

 which is growing in Europe, supposed to have 

 been introduced from America, and described 

 as Anacharis Alsinastrum Babington, re- 

 sembles E. canadensis in habit, but only 

 pistillate flowers have been found, and in these 

 the stigmas are entire. In the North Ameri- 

 can forms with dioecious flowers the staminate 

 sheaths are sessile in the axils of the leaves, 

 and easily overlooked, except in the plant com- 

 mon in the Rocky Mountain region and one 

 specimen from Tennessee, in which the sheaths 

 are peduncled. In the Kocky Mountain plant 

 the staminate flowers are apetalous. 



The subject will be more fully discussed in 

 a paper which Dr. Rydberg is preparing to 

 publish in the Bulletin of the club, as soon 

 as more material has been consulted and cer- 

 tain questions can be answered with more 

 definiteness. 



Both papers were briefly discussed and ad- 

 journment was at 5 :ZQ o'clock. 



C. Stuart Gager, 



Secretary 



THE SCIENCE CLUB OF THE UNIVERSITY Oli" 

 WISCONSIN 



The first regular meeting of the club was 

 held in Science Hall on Tuesday, November 

 5. Dr. Thos. E. Will, secretary of the Amer- 

 ican Forestry Association, delivered an ad- 

 dress on the general subject of forest preserva- 

 tion, with particular emphasis on the proposed 

 forest reserves in the Appalachian and White 

 Mountains. After pointing out the fact that 

 the timber supply of the United States is dis- 

 appearing at an alarmingly rapid rate. Dr. 

 Will showed by a series of significant lantern 

 slides the disastrous indirect effects which the 



removal of the forest-cover produces upon 

 cultivated valleys and slopes, and explained 

 the direct relation between deforestation and 

 the increase in floods. The contrast between 

 western United States, with its large area of 

 forest reserves, and the eastern portion of the 

 country, which has none at all, was clearly 

 brought out by maps thrown on the screen; 

 and the importance of immediately providing- 

 such reserves in the eastern mountains for 

 the protection of the streams which rise among 

 them, was made clear to all. At the close of 

 the lecture a resolution was adopted by the 

 club urging congress to enact a law providing 

 the necessary reserves. 



Eliot Blackwelder, 



Secretary 



DISCUSSION AND CORRESPONDENCE 

 THE EQUATION FOR ONE KILOGRAM OF AIR 



To THE Editor of Science : It is possible 

 that many teachers of thermodynamics may 

 not have noticed that the characteristic equa- 

 tion for one kilogram of air takes the easily 

 rememberable form pv = T/10 when p is 

 measured in standard atmospheres, v in cubic 

 feet, and T in thermodynamic Centigrade 

 degi'ees, the accuracy of the even integer be- 

 ing fully as great as that of the gas law itself. 

 These units are, of course, a curious mixture 

 of the English and continental systems, but 

 this seldom makes much difference in actual 

 problems, and the convenience of the formula 

 for rough mental computations is sometimes 

 very great. 



The data upon which this computation of 

 the gas constant is based are the statements 

 in the third (1905) edition of Landolt and 

 Boernstein that one liter of air under stand- 

 ard conditions weighs 1.2928 grams, and that 

 an American yard is 0.91440 meters, and the 

 value r„ = 273.13° given by Buckingham in 

 the Bulletin of the Bureau of Standards for 

 May, 1907. The value R^=0.1 is consistent 

 with these assumptions within less than one 

 fiftieth of one per cent. 



The corresponding values of Cp and Ov, 

 reduced from the mean of the results of 

 Regnault (1862), Wiedemann (1876) and 



