December, 1921. 



SCIENTIFIC AGRICULTURE. 



131 



to the colour of the individual seeds. In 

 some seeds only one colour occurs, ^^hile 

 in others more than one are present. Yel- 

 low, pink, various shades of purple, and 

 combinations of yellow and other colours 

 are common in practically all samples, 

 while white seeds are only very rarely 

 found. As a rule dark-coloured samples, i.e., 

 samples in which the pink and purple 

 colours predominate, are popularly con- 

 sidered more valuable than light-coloured 

 ones and, this being the case, it is of parti- 

 cular practical interest that the real na- 

 ture of the colour variation is thoroughly 

 understood. 



In order to come to as clear an under 

 standing as possible of the question, the 

 causes of the appearance and the develop 

 ment of the colour in the individual seeds 

 may briefly be referred to. Preyer (1. 

 p. 10) has shown that the colour of the 

 seed is due to the presence of a colouring 

 substance in the palisade cell-walls. 

 Micro-chemical investigations undertaken 

 by him have further revealed that the 

 colouring substance is anthocyan, in the 

 various shades of which evidently are 

 closely determined by physiologico-chemi- 

 cal processes during the ripening of the 

 seed. 



. Considering the nature and physio- 

 logico-chemical development of the col- 

 ouring substance, it is obvious that red 

 clover seed may exhibit different types 

 of colour at different stages of develop- 

 ment. This is especially the case, of 

 course, with such seeds as are, when ripe, 

 richly red or purple-coloured. Thus 

 Scrihaux (1) observed that the seeds 

 that ripen first or, which amounts to the 

 same thing, the biggest seeds are of a 

 darker shade than those that ripen last. 

 A similar observation was made by Mar- 

 tinet (1) in respect to differences in 

 shade in seed from heads on different 

 stages of ripening. Martinet found that 

 in 23 plants of 73, all of which had many 

 heads in different stages of ripening, those 

 heads which were older and consequently 

 contained a larger percentage of fully 

 matured seeds, had seeds of a darker 

 shade than those heads which were not as 

 far advanced. Gernert (1. p. 87) also 

 states that "seeds lighter in colour are 

 frequently found at the tip and extreme 



base of the h<^ad, this being no doubt 

 caused by poor development and imma- 

 turity," ^n assertion substantiated by the 

 observation that "when clover seed is 

 harvested immaturely it shows little or 

 no purple and is usually white or pea 

 colour, which may change to yellow, es- 

 pecially when the seed is cured impro- 

 perly." 



From the above it is evident that, in 

 order to secure reliable evidence as to the 

 colour characteristics of the seed of red 

 clover, it is necessary to examine plants 

 which have fully ripe or, still better 

 dead-ripe seed. Such examinations have 

 revealed that all the seeds in a given plant 

 are of the same general colour. Thus 

 Scrihaux (1) states that the seeds from 

 individual plants present an almost com- 

 plete uniformity and a similar observa- 

 tion is made by Gernert (1. p. 85). Mar- 

 tinet (1) relates t^at in 43 out of 73 plants 

 examined the colour of the seed was uni- 

 form within the individual plants, not- 

 withstanding the fact that the various 

 heads of the individual plants were on 

 different stages of development. Finally, 

 a statement of the writer may be quoted 

 on the subect (Malte, 1. pp. 530-31) as 

 f ollows :" " The author collected, during 

 the past summer, seed from about 250 in- 

 dividual plants of red clover, the major- 

 ity of which Avere growing in British 

 Columbia and western Quebec. In all 

 cases the seed from each individual plant 

 was found to be perfectly uniform as to 

 the general type of colour, although slight 

 modifications ih shade were frequently, 

 but not always present. Although in 

 these investigations no special atten- 

 tion was paid to possible differences in 

 the colour of seeds from heads at a dif- 

 ferent stage of development it might be 

 safely stated that in many cases it is not 

 possible to detect any difference in shade 

 from such heads. This is especially the 

 ease in such plants as have entirely yel- 

 low or entirely purple or violet seed, 

 whereas differences in shade are more 

 often marked in seeds of more than one 

 colour. Even if those graduations in 

 colour really exist, they do not affect the 

 above statement, that all seeds in a given 



