THE PURE LINE HYPOTHESIS. 543 



proved somewhat less suitable for the particular purpose in hand, 

 but nevertheless some results of interest have been obtained, and 

 it is desirable that they should be placed on record. 



The general result with 'both foxg-loves and nasturtiums is, 

 that the '' fluctuating variations " of the Mendelians are inherit- 

 able, and consequently the factors, or group of factors, handed 

 down to the fertilised germ-cell are not constant, but are variable 

 or fluctuating in nature. 



NASTURTIUMS. 



In brief, a dwarf, variegated nasturtium plant (Tropcrolum 

 minus, " Queen ") with red flowers was crossed with a tall 

 climbing, green plant (Tropceohim majus) with pale yellow 

 flowers. The hybrids w^ere reared, and the habit, the characters 

 of the leaves and the colour of the flowers were noted. These 

 hybrids were self-fertilised, and families from each were raised 

 and the same characters were recorded. The plants of this 

 generation exhibited such a strong antipathy to self- fertilisation 

 that a continuation of the experiment became almost imprac- 

 ticable. 



The colour of the flowers which developed during the experi- 

 ment depended on the formation of three colouring matters: (i) 

 A pale yellow substance dissolved in the cell-sap, (2) a red 

 substance also dissolved in the sap, (3) small deep yellow bodies 

 (chromoplasts) imbedded in the cell-protoplasm or lying in the 

 cell-cavity. It appears that (i) is always more or less present 

 in the flower, but (2) and (3) may be absent, or may occur 

 separately or both together in the same flower. Owing to the 

 fact that the deep yellow chromoplasts are unaffected by alcohol, 

 it is easy to detect their presence by immersing the flowers in this 

 fluid. In the course of an hour the pale yellow substance in 

 solution and any red substance that may be present are com- 

 pletely removed, but if the yellow chromoplasts occur in any 

 quantity the flower remains yellow instead of becoming quite 

 transparent. When a flower is not obviously yellow after immer- 

 sion in alcohol the yellow chroinoplasts are said to be " absent." 

 By this it is not implied that chromoplasts are entirely absent, 

 since generally a few may be found by the microscopic examina- 

 tion of the ceils at the base of the petals. 



I. Male or Pollen-plant {A). 



In December, 191 5, there was growing in the garden of the 

 Natal Museum a dwarf nasturtium plant with variegated leaves 

 and rather small medium-red flowers. Chromoplasts absent. The 

 leaves were small and somewhat circular, as is usual with the 

 dwarf variety, and they were conspicuously streaked and dotted 

 with yellowish white. The plant was in a somewhat shaded 

 comer, and did not flower very freely. The flowers were self- 

 fertilised by hand, plugged with cotton-wool, and the petals were 

 then carefully fastened together with minute pins in order to 

 avoid the possibility of foreign pollen entering the flower. 



E 



