J. J. WILLIS — EXPEEIMENTAL GEASS PLOTS AT EOTHAMSTED. 141 



insurmountable difficulties, but "svitli a little practice is easily 

 accomplished. Shai-p lads of about fourteen years of age soon 

 learn (of course, under special training) to distinguish most of the 

 species, even to the smallest fragments of the leaves of any of the 

 grasses. The characters of the leaves of the plants are various, 

 and are found in their texture, surface, colour, point, ligule, mode 

 of curling or folding together vrhen dry, nature of the margin, 

 hairiness, relative prominence of the midrib, etc. By one or more 

 of these characters it is possible to distinguish almost any piece, 

 however small. 



In the Laboratory at HarjDenden is to be seen by the visitor a 

 large upright wall-case, which shows the botanical composition of 

 the herbage on seven selected plots in the twelfth year of the 

 experiments (1867); and at the South Kensington Museum is one 

 contributed by Mr. Lawes, showing the composition in twelve 

 selected plots in the seventeenth year (1872). The quantities of 

 the different plants there exhibited represent the relative pro- 

 portion by weight in which each species was found; and the 

 whole illustrates in a striking manner the domination of one plant 

 over another, under the influence of the different manures applied. 



The general results of the experiments may be briefly summarised 

 as follows : — 



The mean produce of hay per acre, per annum, has ranged on 

 the different plots fi'om about 21 cwt. without manure, to about 63 

 cwt. on the plot most heavily manured. 



The number of s.pecies found has generally been about 50 on the 

 unmanured plot, where there is no marked predominance of one 

 plant over another ; and has been reduced to an average of only 20 

 on the plot most heavily manured, where the effect is to stimulate 

 some of the coarser growing grasses, and other plants, to extra- 

 ordinary growth, and crowd out or otherwise cause to disappear 

 the more weakly species ; and it may be stated as a rule that what- 

 ever the description of manure employed, any considerable increase 

 of crop is accompanied by greater simplicity of herbage. 



Species belonging to the order Graminacese have on the average 

 contributed about 68 per cent, of the weight of the mixed herbage 

 grown without manure, about 65 per cent, of that grown with 

 purely mineral manure (consisting of salts of potash, soda, 

 magnesia, and super-phosphate of lime), and about 94 per cent, 

 of that produced with the same mineral manure with a large 

 quantity of ammonia-salts in addition. 



Species of the order Leguminosae have, on the average, contri- 

 buted about 9 per cent, of the produce without manure, about 20 

 per cent, of that with purely, mineral manure, and less than O'Ol 

 per cent, of that with the mixture of the same minerals and a large 

 quantity of ammonia-salts. 



Species belonging to various other orders have, on the average, 

 contributed about 23 per cent, of the produce without manure, 

 about 15 per cent, of that with mineral manure, and only about 6 

 per cent, of that with the mixture of minerals and ammonia-salts. 



