STATE POMOLOGICAL SOCIETY. I05 



seriously for moisture and materially checking the fall growth 

 of the trees. The gains in nitrogen and humus from these 

 crops, therefore, have evidently not yet compensated for their 

 apparently unfavorable reductions in the moisture supply. 



The gains with the rape, millet and buckwheat are larger than 

 might be expected, and in the case of the latter they may be 

 partly due to a slight advantage in location. Here again, we 

 may note that our results with this class of crops are not mate- 

 rially different from those of Emerson at the Nebraska Station, 

 which were published in 1903 and 1906 in their Bulletins 79 

 and 92. These crops are all frost-killed annuals, though the 

 rape is much more resistant and usually some of its plants sur- 

 vive the winter. As a group, therefore, they offer little or no 

 competition for moisture in the spring, which is apparently 

 much to their credit. Their competition in the fall also, has not 

 been so serious in the present experiment as that of the other 

 frost-killed crops. 



As winter covers, the millet is the best of the present three, — 

 chiefly because of its greater ability to hold the snow, — and the 

 rape is the poorest. The latter usually withers away and dis- 

 appears almost completely during the winter. The buckwheat, 

 also, furnishes but little direct protection to the soil, though it 

 does seem to exert a mysteriously good influence on the physi- 

 cal condition of the latter, making it looser, mellower, and more 

 congenial to moisture. Its general effect, however, is hardly 

 so good as that of millet, and even the latter crop, for the 

 average Pennsylvania conditions, does not yet impress the 

 writer nearly so favorably as hairy vetch, and possibly crimson 

 clover. From present indications, however, and in view of the 

 low cost of their seed, either millet, rape, or buckwheat is likely 

 to prove much more valuable in many cases than many of the 

 plants now sown for orchard covers. 



RESULTS IN ORCHARDS OF EARLY BEARING AGE. 



The next group of results is obtained from orchards ranging 

 from 6 to 20 years old, if we begin with the age of the young- 

 est at the start and finish with that of the oldest at present. The 

 experiments directly concerned in this group are the first three 

 indicated in Table I, and each involves the entire plan shown 

 in Figure i. These experiments were started in 1907, in or- 

 chards already planted, and hence it was not always possible 



