564 THE PURE LINE HYPOTHESIS. 



would be regarded by Mendelians as fluctuating non-inheritable 

 viJriations. Take some character and select the extreme indivi- 

 dual at the two ends of the series, say, the deepest-coloured and 

 the lightest-coloured member. By self-fertilisation raise families 

 from each individual ; then, according to the pure-line hypothesis, 

 the families raised from these two individuals should not differ 

 appreciably from each other : the average colouration in the two 

 families should be the same. If the families are not the same, 

 it implies that the variations in the character are inheritable, and 

 the factors handed down by the darkest and palest individuals 

 differed somewhat from each other. In other words, selection 

 has been eft'ective owing to the fact that the factors are not 

 essentially constant, as has been supi>osed, but are variable, and 

 the bodily expression of the character tends to reflect faithfully 

 any change that has taken place in the factor of the germ-cell. 



To extend the term " mutation" to- any variation which is in- 

 heritable and is readily capable of blending with the normal type 

 is undesirable, and it should be reserved for such inheritable 

 variations as exhibit by the phenomenon of segregation an 

 obvious disability tO' blend intimately with the normal form. 



7. Breeding experiments with foxgloves and nasturtiums 

 have been made in the manner above indicated, and the results so 

 far attained with the foxgloves do not favour the pure-line 

 hypothesis. A similar result has been obtained with nasturtiums, 

 but self-sterility has interfered with the experiments to a con- 

 siderable extent. 



8. A cross was made between two well-known varieties of 

 the garden nasturtium. The seed-plant {Tropceolum majus) 

 was a tall, green plant with yellow flowers, and the pollen-plant 

 {Tropceolum minus, var. " Queen ") was a dwarf, variegated 

 plant with red flowers. These parent plants were homozygous 

 with regard to certain characters — (i) uniform greenness of leaf 

 and variegation of leaf ; (2) tall or straggling habit and dwarf 

 habit. 



9. The colour of the flowers is due to three agents : ( i ) pale 

 yellow substance (chromogen) in solution in cell-sap; (2) red 

 cell-sap (anthocyanin) ; (3) deep yellow chromoplasts. The 

 first-mentioned is always more or less present, while the second 

 and third may be both present together, or either separately. 



10. As a result of the crossing of the two plants 30 seeds 

 were obtained and 21 germinated. Thus nine out of 30 seeds 

 failed to germinate, and this gives a rough relative measure of 

 the sterility resulting from the antipathv to the crossing of the 

 two varieties. 



11. Each of the 21 plants obtained from the cross was self- 

 fertilised, and seeds were obtained. Considerable difficulty was 

 experienced, owing to the antipathy shown to self-fertilisation. 

 The seeds obtained were weak in germinating power, and only 

 5.S per cent, of the seed sown actually germinated. It was clear 

 that the germinating power of the seeds had been reduced by 

 about 30 per cent. 



