It seems probable enough that if Geoffroy had developed his views on evolution he would finally have been led to interpret unity of plan in terms of genetic relationship. But as it was he remained at his morphological standpoint. He did not interpret rudimentary organs as useless heritages of the past; he preferred to think that Nature had prepared double means for the same function, one or other being predominant according as the animal lived in the water or on the land. "To the animal that lives exclusively in the air Nature has granted an organisation suited to this mode of respiration, without however suppressing the other corresponding means, that is to say, without depriving it of a second system which is applicable only to the mode of respiration by the intermediary of water, and _vice versa_."[105] He seems, in one instance at least, to have hit upon the root-idea of the biogenetic law, but he was far from appreciating its significance. He recognised that an amphibian in its development passed through a stage when it was in all essentials similar to a fish, and he saw in this visible transformation a picture of the evolutionary transformation. "An amphibian," he writes,[106] "is at first a fish under the name of tadpole, and then a reptile [_sic_] under that of frog.... In this observed fact is realised what we have above represented as an hypothesis, the transformation of one organic stage into the stage immediately superior." But it is not clear that he considered the development of the amphibian to be a _repetition_ of its ancestral history. He went, however, a certain length towards recognising the main principle of a law which was a commonplace of German transcendental thought, and was developed later by his disciple E. Serres, the law that the higher animals repeat during their development the main features of the adult organisation of animals lower in the scale. Thus he compared fish as regards certain parts of their structure with the foetus of mammals. He compared also Articulates with embryonic Vertebrates in respect of their vertebrae, for in the higher Vertebrates the body of the vertebra is tubular at an early stage of development, and in Articulates the body of the vertebra remains tubular permanently (_supra_, p. 61). As regards their vertebrae, "insects occupy a place in the series of the ages and developments of the vertebrate animals, that is to say, they realise one of the states of their embryo, as fishes do one of the states of their foetal condition."[107] This idea was destined to exercise a great influence upon the development of morphology. A further development of the thought is that certain abnormalities in the higher animals, resulting from arrest of development, represent states of organisation which are permanent in the lower animals.[108]