FIELD CROPS. 1123 



fonnd that a sample from tlu' whole kcriiols j^avc ^2^\.()^ per ccMit of water, wliile 

 tile j,'roiiii(l sample showed only 24.."?() per cent. In another test the whole ker- 

 nels showed '.\~>A>S par cent of moisture and the si'ound kernels 'M.7~t ju'r eent. 

 Sixteen samples S'"ive an average of 20.1.3 per cent for whole kernels and 20.05 

 per cent for fjjround kernels, the moisture content of the different sanijdes vary- 

 inj; from 12.71 i)er cent to .'{."(.(iS per cent. 



The ai)paralus consists of an ('vaporatiuf? chamber with 2 or more c<»mpart- 

 ments, a condenser, and a stand. One hundred yrams of corn are used for the 

 test so that each cuhic centimeter of water in the graduated c.ylinder I'epre- 

 sents 1 per cent of moisture. When the thermometer in the distillation flask 

 registers 1!K)° ( '. the gas is turned off, and the reading of the amount of water 

 e.\i»elled is made 8 or 10 minutes later. The oil used is a good grade of ])ure 

 hydrocarbon oil with a Hash i)oint of 200° to 20.5° C. The oil is poured' into the 

 Hask to first lessen the danger of its being broken by the kernels of corn drop- 

 ping on the bottom. 



Practical suggestions for seed testing, J. J.. Thornber (Arizona Sta. Bui. 

 o'l, i)i>. !)!)-loj, fj(j. /). — An apparatus for testing several kinds of seeds, at the 

 same time is described, and a number of purity and germination tests taken 

 from a series of experiments carried on diiring the last 2 years are given in a 

 table. 



Attention is called to the fact that in 1 sample of alfalfa seed 90.0 per cent 

 was ])ure seed, of which 95 per cent was viable, or 86 per cent of the total 

 sample, while of another "sample only 59 per cent was capable of growing. 

 These 2 samples sold on the market for 15 and 16 cts. per pound, respectively. 

 Two other samples sold for 16 cts. per pound, although 1 contained 29 per cent of 

 inert matter. The samples of rye and barley tested were practically free from 

 inert matter and weed seed, while the wheat contained about 9,000 weed seeds 

 for evei'y bushel. A 30 gm. sample of oats tested contained 10 varieties of 

 weeds, re|)resented by 298 seeds. 



Seed inspection, C. D. Woods and R. L. Hammond {Maine kita. Bui. 13S. pp. 

 50). — The requirements of the Maine seed law are briefly stated, and the results 

 of the examination of samples of seed in 1906 are talmlated in detail.. A list of 

 the weed seeds found in these seeds is also given. 



Pure versus poor seed, II. F. Roberts and G. F. Freeman (Kansa.s ,Sta. kSpec. 

 Cirr.. ./an. .30. 1H07. pp. 21). — This circular compares the cost of obtaining a full 

 stand of alfalfa when pure and poor seed are used, and also reports observations 

 mad(» with refm-ence to blue grass and brome grass seed. 



It is pointed out that in 1905, of the 1,018,206 acres devoted to the culture of 

 forage ci-ops, 602,.5r>0 acres was in alfalfa, and that the cost of seeding this 

 acreage at the rate of l.~> lbs. per acre with seed at K! cts. i)er pound, every seed 

 being good, would amount to .$1,446,128, while taking as a basis from among the 

 number of alfalfa seed sam]»les analyzed by tlie station 28 showing an average 

 of total iini)urities of 46.1 per cent and an average number of seeds true to name 

 but incajtable of germination of .■',4.."') jier cent, it would have cost .$1.9:>.5,042 to 

 secure a full stand on the same area. The pi'incipal impurities found are dead 

 and defective alfalfa seed itself, trefoil. English plantain or buckhorn, doddei*, 

 Russian thistle, crab grass, foxtail, and other weed seeds, and the adulterants 

 are tr(>foiI, bur clover, and sweet clover. 



In ;i certain lot of seed analyzed at the station only 20.2 per cent was jmre and 

 i;tp;ible of germinating. In order to obtain from the use of this pure si-ed as 

 much of a stand as could have been se(;ured from 15 lbs. of standard seed aver- 

 aging 83 per cent of the seed true to name and capable of germinating, 73.9 

 lbs. would have been necessary, and the cost of seeding an acre would have been 

 brought up to $11.92, as compared with $2.40 for the standard seed. The 28 



