F. KiDT) AM) (\ Wkst llM 



produced, but we have been able to trace lew critical published results 

 bearing upon this point. 



The results of investigations by J. L, Jensen (li') indicate tliat the 

 average grain-weight of cereals is clearly rslated to climatic conditions. 

 In brief, the average grain-weight decreases as the continental character 

 of the country becomes more marked and, vice versa, it increases with an 

 insular or coast climate, and in connection with such climate it also 

 increases with rising temperature. In the following table the average 

 weight in gram? of 10,000 grains is given for wheat, barley, oats, and 

 rye resportivolr from each country investigated: 



Tahlk Ua. 

 (rrain-ireighls accordiiu/ to climatic cotulifion.s. 



Average weight in grams of 10,000 grains. 



Such records as the following, however, occur. Raulin(60) grew 

 varieties of wheat at Lyons and in the Ardennes for comparison. The 

 crop ripening at Lyons was planted the following year in the Ardennes, 

 and vice versa, for several years. The conclusion drawn was that "les 

 poids de 100 grains paraissent tendre a Lyon et dans les Ardennes 

 d'annee en annee, vers des limites differentes : inf erieure a Lyon (Pierre- 

 Benite), superieure dans les Ardennes." 



The interesting indications provided by Combes' work that environ- 

 mental conditions which result in a weak or a slow development of the 

 parent plant may be correlated with the production of larger seeds though 

 fewer in number, find confirmation in a series of statistical studies by 

 Waldron(75) and Harris (39). Waldron (I.e.) found in the case of oats a 

 strong correlation between short culms, short heads and small number 

 of grains on the one hand, and heavy-seededness on the other. On the 

 basis of some data of Lyon's he worked out for wheat a similar, but less 

 marked, negative correlation between the number of grains per head and 



