May, 1922 



SCIENTIFIC AGRICULTURE. 



<'()lumn ill the tables and come to their 

 own ronehision is to the relative milling' 

 value and baking qualities of tlie wheats 

 tested. 



In the Birchard tables no flour yields 

 have been i-eported. This isi due to tli-e 

 fact that only five pound samples Avere 

 sent to be tested and it is considered im- 

 possible to obtain accurate results on an 

 experimental mill, without duplicating 

 the testsi many times. 



In these tables tlie wheats have been 

 arranged according to their baking quali- 

 ties as near as can be estimated. Again 

 it must be understood that the arrange- 

 ment is not considered final and absolutely 

 correct. We are entirely open to criticism 

 in this respect and hope that those who 

 study the tables will pay due regard to 

 each item and come to their own conclu- 

 sions Avith regard to the relative perform- 

 ance of the different wheats. 



In a comparison of yields of varieties it 

 is always found advisable to average the 

 yields for a number of years in order to 

 obtain accurate results. It "was at first 

 thought that it would be possible to aver- 

 age our milling and baking test data in 

 much the same manner and thus prepare 

 summary tables Avhich would indicate the 

 actual milling and baking values of the 

 different Avheats very accurately. On 

 further consideration, however, it Avas 

 found that this method Avoiild be more or 

 less impracticable and Avould be likely to , 

 introduce scA^eral inaccuracies. The rea- 

 sons for this may he set doAvn as f oIIoavs : 



1. — Much important data on both kinds 

 of tables are giA'eii in legends and not in 

 figures. This sort of data could not be 

 averaged, and of omitted, the milling and 

 baking qualities of the Avheats under con- 

 sideration Avould not be truh'' represent- 

 ed. 



2. The data from the tAA^o sources are 

 presented in A-ery different form. The 

 tables are thus not truly comparable and 

 at the most only four year aA'erages could 

 be obtained. 



3. Texture, color, and shape in the A'ery 

 poor (juality Avheats are not giA-en a score 

 and accurate results could not be obtain- 

 ed if they Avere omitted. 



4. To possess high milling and l)akiiig ■ 

 quality a Avheat must be good in every 

 particular and consistenth" so. Certain 



varieties go very low one yeai' and hio-h 

 ni another. Certain other varieties dur- 

 ing .some years will drop very low in one 

 point and in another point durino- other 

 years. In both of these, averaging^brings 

 the Avheat up closer to normal and er- 

 roneous results are obtained. 



To conclude : the summarizing of mill- 

 mg and baking test data is not usuallv 

 praetieable; therefore, the only sati.s- 

 tactory Avay to arrive at conclusion Avith 

 regard to the relative milling and baking 

 values of the different varieties is bv a 

 careful perusal of the results bv ve'ars 

 following out closely the shoAA-ing ^nade 

 by each variety each year. 



Having carried out this program Ave 

 have found than it is practicallv impossible 

 to arrange the varieties definitelv in order 

 according to their milling and bakino- 

 values. They can, hoAvever, be quite con- 

 veniently arranged in different groups 

 each group having in common some par- 

 ticular qualities AA-hich distinguish it from 

 the rest. 



We noAv propose to set doAvn these 

 group.^ m order and discuss the milling 

 and baking values of the varieties AAiiicfi 

 compose them. 



Group 1.— Marquis, Red Fife, Kitchen- 

 er, Red Bobs. 



In long continued tests Marquis has 

 proven itself to be a Avheat of excellent 

 milling (fuality. During the first three 

 years that the tests Avere carried on it ap- 

 peared to be slightly inferior to White 

 Bobs, a Avhite Avheat, and it is to be re- 

 gretted that the tests on the latter Avere 

 discontinued from that time on. In other 

 years Marquis has headed the list in the 

 majority of cases- and has alAA^ays been 

 very close to the top. Its excellent per- 

 formance is due in great measure to Jie 

 consistency Avith AA^hich it gives results. 



Red Fife Avheat has been tested along 

 Avith Marquis for a period of eight vear.s". 

 In two cases 1914 (Table II) and" 1917 

 (Table V) it Avas slightly .superior to 

 Marquis but in neither case Avas the dif- 

 ference marked. In other years it has 

 fallen beloAv Marquis and in some years 

 has given rather Ioav tests. There is no 

 (juestion, hoAvever, as to its being a good 

 milling Avheat, the poor tests probablv 

 heing due to the nature of the samples 

 themselves. In 1916 (Table IV) the loaf 



