776 EXPERIMENT STATION RECORD. 



A thermostat, Bkim.kmont (/>'«/. Soc. Chim. Paris, lS-14 {18D5), No. 5,j)p. 22S,SW, 



fig- ])■ 



An apparatus for grinding peanuts, J. G. W. Kemm (Nenc Zlschr. liiihenz. Ivd., 

 S4 {1895), No. G,2)p. 70, SO, pi. 1). 



Some new pieces of chemical apparatus, P. Nyssens {Rev. Ayroi). Louvain, S 

 {1S94}, Xo. 4, pp. S45-S52, fujs. 4). — Descriptions of fat extractor, mechanical agitator, 

 and water bath with constant level. 



BOTANY. 



The method and application of a quantitative botanical anal- 

 ysis of meadows, A. Voigt {Landw. Jahrb., 23 {1894)., No. 4 and 5, pp. 

 707-78ti, figs. 2). — The author reviews the methods heretofore employed 

 in making' botanical analyses of meadows, and points out some of their 

 deficiencies. The value of botanical analyses as compared with chem- 

 ical analyses is shown. Chemical analysis of grasses does not always 

 represent the true value of meadows either for pasturage or hay as 

 well as a botanical analysis, as from the latter may be seen the relative 

 value of various species, based on the avidity with which stock will eat 

 them. 



The method suggested by the author for securing representative 

 material for a quantitative botanical analysis of an area of 1 are is 

 as follows : Immediately after mowing from each swath at distances 

 of about 50 cm., specimens are taken in proportion to the size and 

 abundance of the material. This forms what is called type 1, and will 

 weigh from 17 to 30 kg., depending on the abundance of the material. 

 This type is then spread about in a circle of 1.5 to 2 meters in diameter 

 until it is of a uniform thickness and of apparent homogeneity. From 

 this circle type 2 is taken, by removing a handful at a time from differ- 

 ent parts of the circle until about one fourth of the total quantity is 

 secured. This type is similarly treated to secure types 3 and 4, the 

 final one of which will weigh from 2 to 4 kg. From this last so-called 

 type the botanical determinations are made, a part being retained for 

 control analysis. When strong plants, as thistles, meadow sweet, etc., 

 are present, they must be broken up, so as to permit of a more eveu 

 distribution of their material. The principal difficulty in such an 

 analysis will be with the grass leaves, such as are found in aftermath. 

 These must be removed, assorted, weighed, and their proportion to 

 the whole ascertained. Their determination is difficult, and should be 

 made by a specialist. The same is true of the sedges, mosses, etc. In 

 this way the species of plants found in a meadow and their relative 

 abundance may be determined with reasonable accuracy. 



The author divides the more common grasses into 4 categories, the 

 groups and species being arranged according to their relative value for 

 forage, the first being considered the best, the others decreasing in 

 value according to their position in the list. Group 1, Festuca elatior, 

 Loliuni perenne, JJactylis glomerata, Arrhenatherum elatius, Avena flaves- 

 cens, Phleum pratense, Foa pratensis, F.trivialis., Agrostis alba, Glyceria 

 Jtuita7is, G. aquatica, and Cyno.^urus cristatus. Group 2, Festuca rubra, 



