experiments upon the yellow day lily, plants impregnated from their own stamens cannot be made to set seed at all. 'So near,' says his able successor, Hermann Müller, 'was Sprengel to the distinct recognition of the fact that self-fertilisation leads to worse results than cross-fertilisation, and that all the arrangements which favour insect-visits are of value to the plant itself, simply because the insect-visitors effect cross-fertilisation!' As in most other anticipatory cases, however, it must be here remarked that Sprengel's idea was wholly teleological: he conceived of nature as animated by a direct informing principle, which deliberately aimed at a particular result; whereas Darwin rather came to the conclusion that cross-fertilisation as a matter of fact does actually produce beneficial results, and that therefore those plants which varied most in the direction of arrangements for favouring insect-visits were likely to be exceptionally fortunate in the struggle for existence against competitors otherwise arranged. It is just the usual Darwinian substitution of an efficient for a final cause. Even before Sprengel, Kölreuter had recognised, in 1761, that self-fertilisation was avoided in nature; and his observations and experiments on intercrossing and on hybridism were largely relied upon by Darwin himself, to whom they suggested at an early period many fruitful lines of original investigation. In 1799, again, Andrew Knight, following up the same line of thought in England as Sprengel in Germany, declared as the result of his close experiments upon the garden pea, that no plant ever fertilises itself for a perpetuity of generations. But Knight's law, not being brought into causal connection with any great fundamental principle of nature, was almost entirely overlooked by the scientific world until the publication of Darwin's 'Origin of Species,' half a century later. The same neglect also overtook Sprengel's immensely interesting and curious work on fertilisation of flowers. The world, in fact, was not yet ready for the separate treatment of functional problems connected with the interrelations of organic beings; so Knight and Sprengel were laid aside unnoticed on the dusty top bookshelves of public libraries, while the dry classificatory and systematic biology of the moment had it ail its own way for the time being on the centre reading-tables. So many separate and independent strands of thought does it ultimately require to make up the grand final generalisation which the outer world attributes in its totality to the one supreme organising intelligence.