254 DR J. M. MACFARLANE ON THE 



But that the crossing of species which have diverse colouring, even though nearly 

 related, often produces profound molecular disturbance and instability in the offspring, was 

 early recognised by Darwin, and has repeatedly been insisted on since. Thus Darwin 

 obtained from a cross of the red and blue Anagallis mongrel progeny, some of which were 

 red, some blue, and some intermediate in colour. This may be accounted for by 

 supposing that a state of instability had been brought about by an overstraining of 

 that characteristic elasticity of the protoplasm, which in most cases of species-crossing 

 gives an evenly blended product. 



That there may be morphological peculiarities associated or correlated with colour 

 production, and entirely confined to one parent, is clearly shown by the account given of the 

 large se}3al of Cypripedium Leeanum. With a copious production of a deep crimson 

 pigment in C. insigne, aggregated into large sharply-defined spots, there is an absence 

 of the simple hairs that are abundantly present over the rest of the inner sepal surface. 

 The other parent has many gland-tipped hairs over the areas, which are uniformly white 

 in it, but spotted in C. insigne. Now in C. Leeanum, while the pigment spots are 

 undoubtedly much paler in colour, their presence always points to an entire absence of 

 hairs. This, along with other facts shortly to be reviewed, points not only to a molecular 

 instability in the hybrid tissues, but to an ultimate predominance of one sexual element 

 over the other, alike in the formation of chemical products and in the upbuilding of 

 permanent tissues. 



We shall see later on that the petals of the remarkable graft hybrid Cytisus Adami 

 are usually exactly intermediate in colour between those of the yellow and purple species 

 with which they are associated on the same tree, even though many of the tissues take 

 entirely after one or other parent in a mixed manner ; but instances appear to be not 

 unfrequent * where half or some part of a petal is like one parent, the remainder like the 

 other. Here again is one-sided sexual predominance in colour formation. 



The evidence at present to hand warrants the assumj)tion that the majority of plant 

 hybrids are exactly intermediate between the parents in colour production if the colours 

 readily blend, but that some show a greater or less degree of instability, which passes 

 by transition cases to complete resemblance to one parent. 



III. 

 Comparative Chemical Constitution of Hybrids and of their Parents. 



This subject might appropriately enough have been dealt with under the last 

 head, while my observations are very fragmentary. I give them as they are, in the 

 hope that thereby attention may be drawn to what is a wide and important though 

 difficult field for research. 



On placing twigs of Cytisus Laburnum, C. purpureus, and the intermediate graft 

 hybrid, or C. Adami, in separate bottles of alcohol for preservation, I was surprised soon 



* Dakwin, Animals and Plants, vol. i. p. 414. 



