MATERIALS AND METHODS. 



11 



watermelon (Sweet Mountain) were grown in Michigan. The corn 

 "B" (Minnesota Sweet), was grown in Nebraska, the cabbage (Win- 

 ningstedt), in Washington, and the lettuce (Black-Seeded Simpson), in 

 California, while the radish (Early Scarlet Turnip-Rooted), carrot 

 (Chantenay), pansy (mixed), and Phlox drummondii (mixed) were 

 grown in France. The seed was all of the harvest of 1899 and was 

 received at the botanical laboratory of the University of Michigan on 

 January 27, 1900. 



On January 30, 1900, germination tests were made, showing the 

 vitality of the seeds to be as follows: 



Vitality of seeds tested January SO, 1900. 



GERMINATION TESTS AND APPARATUS. 



In the preliminary work several methods of testing were tried, but 

 as none proved as serviceable as the "Geneva tester," this apparatus 

 was adopted for all subsequent tests as recorded in the following 

 pages. The detailed construction of this tester need not be described, 

 for it is simple and quite familiar to all. However, some modifications 

 were made in the preparation of the apparatus, and some precautions 

 taken in the manipulation, which have proved to be of much value. 

 The brass wires originally and ordinarily used to support the folds of 

 cloth were replaced by glass rods of 6 to 7 mm. diameter. Rods of 

 this size are much heavier than is necessary to support the folds of 

 cloth, but the chief advantage in having rods of large diameter is that 

 in case of the germination of large seeds the folds can be drawn near 

 together at the top and still have sufficient space within the fold for the 

 seeds. On the other hand, in the germination of small seeds that 

 require considerable quantities of air, the folds can be closed at the 

 top by bringing the rods together, thus insuring more uniform condi- 

 tions throughout the fold and at the same time leaving sufficient space 

 above the seeds for an abundant supply of air. The chief advantage 

 in substituting glass rods for brass wires is in removing the possible 

 source of injuiy resulting from the poisonous action of the dissolved 

 copper. 



Another error frequently, if not always, made in using such a tester 

 is in allowing the ends of the cloths, or sometimes the bottoms of the 



