24 



Richard H. Bocrker 



case of certain Pacific coast species four months was apparently 

 necessary. Three months ordinarily is plenty time enough ; 

 usually if a seed in the forest fails to germinate in that time, it 

 usually does not germinate at all, especially in the west where 

 the dry period sets in after the spring is over. 



Ten of the species mentioned in the foregoing tables failed to 

 germinate. These species were Larix europea, Acer saccliarum 

 (both), Liriodendron tulipifera, Tax odium distichmn, Liqui- 

 damber styraciflua, Fraxinus americana, all of which were supplied 

 by commercial seedmen. If the data regarding the collection of 

 these seeds is bona-fide, their failure to germinate must be ex- 

 plained by the fact that they had not completed their resting 

 period. In the case of Juniperis virginiana, Gleditschia tria- 

 canthos, and Pinus monticola, whose place and date of collection 

 is known absolutely there can be very little doubt as to why they 

 failed to germinate. 



The soil used in all cultures (except the sand and gravel) was a 

 garden loam of excellent quality with a mixture of about 25 per 

 cent, of white sand. The mixture was prepared in the green- 

 house. This made a very good soil for experimentation purposes. 

 The sand used was common white, quartz sand with bin a very 

 small per cent, of hornblende and magnetite. The gravel was the 

 kind used by the large construction companies around Lincoln 

 for concrete work. Mechanical analyses of representative 

 samples of these soils are given elsewhere. 



All seeds were planted in rows at a depth which was 2^2 times 

 the shortest diameter of the seed as near as this was determinable 

 by the unaided eye. The rows averaged 3 inches apart and about 

 24 inches in length. In general 200 seeds were used of each 

 species when the seeds were of medium size or smaller ; for some 

 of the western pines only 100 were used because of their large 

 size. 



The Control of Habitat Factors 



As has been pointed out, the only safe way to study the effect 

 of the factors of the habitat upon the life of the plant is to 

 measure one variable factor while all the rest are kept constant. 



