44 JOURNAL OF THE ROYAL HORTICULTURAL SOCIETY. 



at the intermediate altitudes. Expressing this result— which I believe 

 to hold for a large number of flowers — in a rough and ready way, it 

 may be said that Low Sun promotes yellows, Middle Sun reds, and 

 High Sun purples, in our latitudes and climate. And this has been 

 found to be true for seasonal as well as diurnal colour-changes. Here 

 are many specimens illustrating this, and you can see that removing 

 middle and high sun from a flower in summer increases the yellow 

 colour in the same way as low sun does in autumn or early spring. 

 I shall not enter into the question now v/hether this is due to the 

 yellow pigment being formed or only unmasked. For the first four 

 years of these experiments with Nasturtiums, only South African seed 

 was used, but it was then thought desirable to introduce some seed from 

 Scotland for comparison.* Proliferation, such as could be seen in the 

 flower exhibited, of the Rose ' Carmine Pillar,' and believed to have been 

 entirely due to sunlight, was the first important phenomenon to take 

 place in the Nasturtiums under the stimulus of sunlight. It appeared 

 simultaneously in three plants, two of which were grown from South 

 African and one from Scotch seed. It was also a Scotch plant which 

 showed the first great structural change associated with a change of 

 colour when the flower developed three spurs instead of one. This 

 occurred in 1912; and it has now become so common that such flowers 

 are to be found in any aspect, without any artificial screening. Here 

 are several specimens out of over sixty which occurred in the open 

 garden in 1914. You will see that many important structural 

 alterations accompany this growth. The number of petals varies, and 

 instead of the usual two sessile and three unguiculate petals, four out 

 of the five may be sessile, or the number may be increased to six — all 

 of which may be sessile. In cases like the last, the margins of the 

 petals sometimes adhere ; and when such a flower has three spurs, as 

 you see this one has, we are instinctively reminded of the Aquilegia, 

 which belongs to another order, the Ranunculaceae. I would draw 

 your attention to these unusual coloured parallel veins in the un- 

 guiculate petals, which are found to denote that the petal is nearly 

 sessile in character. It is significant of what is going on at the base of 

 the lamina. You will notice also that with these changes the sub- 

 tending leaves disappear from the nodes, and the growth at the inter- 

 nodes is so abnormal that six and seven flowers will spring together, 

 like an umbel, from a lateral shoot not an inch long, or from the same 

 length of the axis itself. The whole plant appears to be affected. 

 Are we here approaching the inflorescence of the Geraniaceae ? There 

 are numerous cases of fasciation of the spur, and as many as four 

 have been found adhering laterally. The experiments have lasted too 



* This was the first outside seed admitted into the same garden as the 

 experiments, and by this time the changes of colour with sun's altitude were 

 unquestionable. Seeds are either sown in boxes and the plants bedded out, or 

 they are sown in paper bags, with specially sifted earth, in the beds they are to 

 occupy. By preventing any possible mixture of seed from individual plants, 

 any sudden change of colour in the petals or of growth in an organism was 

 detected. When the lessons learnt with the Pretoria seed were applied to that 

 obtained from Scotland the same results followed. 



