WEED SEEDS IN THE SOIL. 371 



but eight inches deep. The total number of weeds observed on this field 

 included 107 species. 



The -wTiter made investigations, during the past five years, along similar 

 lines. In addition to ascertaining the number of viable weed seeds in the 

 different samples of soil used, it was also the intention to learn what difference 

 there was in this respect, between soils carefully tended and those receiving 

 poor cultivation. Still another object was to learn at what depth most of 

 the seeds are usually found, and also what species are most prevalent. 



Samples of soils were secured from various sites, during the months of 

 November and December, and placed in boxes or pots in the greenhouse. 

 When the growing weeds were large enough to allow definite determination of 

 the species they were pulled and recorded. This process was continued until 

 all viable seeds had germinated. The soil was stirred occasionally or repotted 

 to hasten germination. All samples were kept for a considerable period after 

 the last specimens of weeds were removed to make sure that all viable seeds 

 had opportunity and time to grow. 



In each of the following five samples a square foot of soil was taken to 

 a depth of six inches and separated into the upper and the lower three-inch 

 layers. These samples were secured from the following sites: 



1. A rye field, which prior to this time had been repeatedly planted to 

 corn for several years, the crops having been rather poorly tended. (Two 

 samples.) 



2. An old abandoned orchard, used to some extent for gardening purposes. 



3. An experimental plot on Purdue Farm, used mostly for corn, cow- 

 peas and soy beans. This ground was carefully tended. 



4. A garden patch, well tended until midsumn^er, then allowed to run 

 to weeds. 



5. A blue-grass sod, in an old i)ermanent pasture. 



6. In addition to the above a cubic foot sample was taken from site No. 

 1 (the rye field), and separated into the successive inch layers, each layer 

 being potted separately. 



The first specimens of weeds were picked, in each case, about a month 

 after the soil samples were potted and placed in the greenhouse. The pulling 

 process was continued, al)out two months apart, for nearly two years. 



Table 1, on the following j^age, shows the number of plants of each species 

 that grew from the upper and the lower three-inch layers of the various soil 

 samples. 



The figures in the tal)le indicate that the grass-weeds are usually the most 

 common intruders in the gardens and fields. This is especially true of crab- 

 grass, whicli undoubtedly is the most common and persistent grass-weed in 

 the state. 



The comparatively large numbers of peppergrass, shepherd's purse, and 

 carpetweed, found in the garden, justify the title of "social weeds," fre- 

 quently applied to these particular species. They do not seem to be able 



