COMPARISON OF SEED TESTING 691 



to year, and the depth of covering, method of watering and sifting, 

 temperature, and air circulation must all be identical. 



If after extreme care the sand method of testing should be standard- 

 ized, could the results obtained be corroborated or duplicated by future 

 workers ? The writer does not believe that this may be done, for there 

 are too many variables involved. And here we have come to the main, 

 reason why so little i)rogress has been made in testing of coniferous 

 seed; there has been no standardized method or equipment. 



Foreign seed investigators have long since abandoned the soil tests. 

 One of the foremost Austrian investigators, Wilhelm Kinzel,^ states in 

 this connection that a critic who demands that comparative germination 

 tests be carried on in soil appeals to such practitioners as have never 

 had to do with the technique of comparative germination tests, and 

 gives as the main reason the fact that a normal soil does not exist. 



It was for the purpose of overcoming the disadvantages attending 

 the sand tests and with a desire to perfect an apparatus which would 

 give more uniform results, more rapid germination, and which could 

 be more readily standardized that germination tests were begun at the 

 Priest River Experiment Station with different media and containers. 

 Among the methods tried were petri dishes, a water-jacket germinating 

 chamber, an incubator, and a tank containing water over which the 

 seed was suspended in folds of cotton flannel which touched the water. 

 All these methods have been discarded. Since it would require too 

 much space to give the results in detail, it is sufficient to state here that 

 the main disadvantage of all was the impossibility of regulating heat 

 and moisture conditions to suit individual species. Some of the seed 

 germinated well for a while and then rotted, while some failed entirely 

 to germinate. These objectionable conditions were largely overcome 

 in experiments with a germinator modeled after the one in use at the 

 Danish Seed Control Station in Copenhagen, known as the Jacobsen 

 germinator. 



The Jacobsen germinator is in itself a product of evolution in seed 

 testing. It was perfected by Miss Ingeborg Jacobsen, of Copenhagen, 

 after years of experience and visits to most of the European seed-test- 

 ing establishments. The one used at the Priest River Experiment 

 Station consists of a galvanized sheet-iron tank 13 by 22 by 5 inches, 

 slightly wider at the top than at the bottom. The top is made of glass 

 strips which rest upon an inside rim a half inch below the upper edge. 

 The glass strips are placed close together and have notches ground 



'Zeitsclirift f. Fnrst u. T.aiulw., Vol. i.^, April and May. 1013. 



