334 LAWS OF VAEIATION. Chap. XXYI, 



" tion of distinct organs occurs, these are formed in continuity 

 " with, each other." He adds that organs already differenti- 

 ated probably in no case become united to homologous ones. 

 M. Dareste does not speak* quite decisively against the law of 

 8oi pour soi, but concludes by sa3^ing, " On se rend parfaitement 

 " compte de la formation des monstres, si Ton admet que les 

 " embryons qui se soudent ajopartiennent a un meme ceuf ; 

 " qu'ils s'unissent en meme temps qu'ils se forment, et que la 

 " soudure ne se produit que pendant la premiere periode de la 

 " vie embryonnaire, celle ou les organes ne sont encore con- 

 " stitues que par des blastemes homogenes." 



By whatever means the abnormal fusion of homologous 

 paits is effected, such cases throw light on the frequent 

 presence of organs which are double during an embryonic 

 period (and throughout life in other and lower members of the 

 same class) but which afterwards unite by a normal process 

 into a single medial organ. In the vegetable kingdom Moquin- 

 Tandon^ gives a long list of cases, showing how frequently 

 homologous parts, such as leaves, jDctals, stamens, and j)istils, 

 flowers, and aggregates of homologous parts, such as buds, 

 as well as fruit, become blended, both normally and abnor- 

 mally, with perfect symmetry into one another. 



The Variabilit]/ of Multiple and Homologous Parts. — Isidore 

 Geoffrey^ insists that, -vvhen any part or organ is repeated 

 many times in the same animal, it is particularly liable to 

 vary both in number and structure. With respect to number, 

 the proposition may, I think, be considered as fully estab- 

 lished ; but the evidence is chiefly derived from organic 

 beings living under their natural conditions, with which we 

 are not here concerned. Whenever such parts as the vertebn^ 

 or teeth, the ra^'s in the fins of fishes, or the feathers in 

 the tails of birds, or petals, stamens, pistils, or seeds, are very 

 numerous, the number is generally variable. AVith respect to 

 the structure of multiple parts, the evidence of variability is 

 not so decisive ; but the fact, as far as it may be trusted, 



*' Archives de Zoolog. Exper.,' Jan., ^ 'Hist, des Anomalies,' tona. iii, 



1874, p. 78. pp. 4, 5, 6. 



* ' Tevatologie Veg.,' 1841, livre iii. 



