1 82 Journal of Agricultural Research vo1.xix,no.4 



by the Stutzer method. In badly frosted wheats he found that from 

 lo to 1 6 per cent of the nitogen was in the nonalbuminoid form. There 

 was practically no difference in the relative amounts of these constituents 

 in the flours milled from these wheats, a fact which will be discussed 

 later in this paper. 



Analyses for the crude food constituents, conducted along the con- 

 ventional lines, have brought out several facts. Frozen wheat runs 

 higher in fiber, ash, and crude fat than sound wheat, although the differ- 

 ences are not always great. The carbohydrate content of frozen wheat 

 is lower than that of sound wheat. The total nitrogen content may be 

 higher or lower, depending probably on some other factors. The mois- 

 ture content varies with storage conditions but is undoubtedly greater in 

 frozen wheat at the normal time of cutting than in sound wheat. 



EXPERIMENTAL WORK 



In order to obtain samples of sound and frozen wheat of the same 

 varieties and grown as nearly as possible under the same conditions, 

 plots were seeded at intervals of a week, starting at the beginning of the 

 normal seeding period and ending about two months later. This insured 

 the likelihood of securing samples frozen at different stages of growth. 

 Marquis wheat was used in the experiments discussed in this paper. A 

 series of plots was seeded in 191 7, beginning May 12 and ending June 30. 

 One i/4oacre plot was seeded each week during the interval, making a 

 total of eight plots. The same procedure was followed in 191 8, starting 

 April 29 and ending June 18. Two series of samples were thus obtained. 

 The soil used was a black sandy loam on the grounds of the Montana 

 Agricultural Experiment Station. All plots were irrigated in the middle 

 of July and obviously were not at the same stage of development when 

 irrigated. In all plots the wheat was cut either shortly after maturity 

 or immediately after the first killing frost. 



It will be noted that in each series of samples only the last two plots 

 seeded were badly frozen. In the first series the plot seeded last was 

 severely frozen when in the late milk stage, and in the second series the 

 wheat from the corresponding plot was less severely frozen when in the 

 early dough stage. In the two most severely frozen samples a large per- 

 centage of the kernels were green, shrunken, and "blistered." These 

 two samples may be considered to represent very extreme cases, and such 

 instances are likely to occur only under the most exceptional climatic 

 conditions. The plots seeded next to the last ones are probably more 

 typical of conditions which are likely to occur in actual farming practice, 

 the one in the 191 7 series being more severely frosted than that in 191 8. 

 Although it is difficult to measure the exact degree of frosting or freezing 

 in a given sample of wheat between certain limits, these samples present 

 an appearance quite similar to the majority of frosted wheat samples 



