De Vries has started from the phenomenon above mentioned, so well known to nursery-gardeners and which is often of considerable financial importance to them; the phenomenon that suddenly in some of their flower-beds single variations appear which are constant when reproduced by seed. By systematically propagating these exceptional specimens a quantity of seed may be obtained in a few years that can be brought into the market. Soon, however, the variety loses its mercantile value; seed may now be obtained by anybody in increased quantities from the plants that have already been sold.
Are there any of our native species of plants in which the same phenomenon produces itself naturally? was the question which de Vries set himself. And if so, can they teach anything about the formation of species? In commencing the inquiry he started with about a hundred different species. Of all these, only one exhibited the property sought for—this, however, in such a way as to throw full light on the subject in most unexpected directions.
The species of plant which de Vries actually managed to detect in the act of 'mutation' on certain fields in Graveland, and which has continued to do so with perfect distinctness during many years in the Amsterdam Botanical Garden, bears the name of Œnothera Lamarckiana. It is one of three species of the genus that have been brought over from the United States and is now running wild in Europe.
De Vries has thus convinced himself that the great majority of plants about us do not show cases of 'chance variations,' 'mutation variation' per saltum; in other words, that the species that have been observed for many centuries may be said to be stable, invariable. But they are stable in so far only as—perhaps with very long pauses—periods of mutability appear, during which, next to the stable central species, new sub-species appear that are also stable when propagated by seed. Further experiments, however, are required to throw light on the periodicity.
Another conclusion was this, that the species which do produce mutations bring forth not a single mutation, but quite a number of them, varying among themselves. The mutations occur both in plants growing in the wild state and in those samples of Œnothera Lamarckiana that are bred under supervision. Their frequency, as determined by exact statistical tables drawn up by de Vries, varies between one and two per cent. Of the 50,000 Œnothera which de Vries has observed during ten years' culture, there were 800 that could not be designated by the name Œnothera Lamarckiana.
A somewhat skeptically disposed person may claim that this number of 800 indicates the number of the most marked deviations which were noticed among 50,000 plants; that, in other words, they are individual, fluctuating variations that would also be found in the same