311 



grasses, especially the pasture grass, has been studied at the Pennsyl- 

 vania State College Agricultural Experiment Station, by G. L. Holter 

 (Agr. Sci., Vol. Ill, p. 285), in connection with studies on the 

 yield per acre. Samples of grass were cut every few days during the 

 season (of 1887), but the comparisons with rainfall and temperature 

 showed no definite relation, except, perhaps, that the percentage of 

 ash increased as the temperature diminished. The following table 

 gives the figures showing the average rainfall and temperature from 

 the middle of one period to the middle of the next, and for the average 

 of seven plats of ground : 



If we assume that the whole season extended from May 10 to Sep- 

 tember 29, we may compute the average daily growth, which will be 

 found to be very large at first, but rather uniform from June 13 to 

 September 16, after which it steadily diminishes. The irregularities 

 in the growth from one week to the next have no simple connection 

 with rainfall or temperature, but there is no evidence to show that 

 other climatic elements, such as sunshine and evaporation, would not 

 have thrown some light upon the subject. 



• Mr. Holter has also experimented on the yield per acre of pasture 

 grass, as measured week by week during the growing season of 

 1888 and 1889 at the Pennsylvania State College Agricultural 

 Experiment Station. (See Agr. Sci., Vol. V, p. 52.) The plat 

 experimented upon represents an average of the uplands of the Alle- 

 gheny Mountains. The weather of the season was most favorable 

 for the growth of grass, having a heavy, evenly distributed rainfall. 

 The following table shows the dates of cutting and the average daily 

 growth between these dates, expressed in pounds of dried grass per 

 acre. It will be seen that there was a rapid increase in growth up 

 to May 21, after which there was a steady decline. The daily 

 average for the whole season of one hundred and seventy-eight days 



