We now come to cases closely analogous with those just {136} given, but different, inasmuch as individual plants alone of the species are self-impotent. This self-impotence does not depend on the pollen or ovules being in a state unfit for fertilisation, for both have been found effective in union with other plants of the same or of a distinct species. The fact of these plants having spontaneously acquired so peculiar a constitution, that they can be fertilised more readily by the pollen of a distinct species than by their own, is remarkable. These abnormal cases, as well as the foregoing normal cases, in which certain orchids, for instance, can be much more easily fertilised by the pollen of a distinct species than by their own, are exactly the reverse of what occurs with all ordinary species. For in these latter the two sexual elements of the same individual plant are capable of freely acting on each other; but are so constituted that they are more or less impotent when brought into union with the sexual elements of a distinct species, and produce more or less sterile hybrids. It would appear that the pollen or ovules, or both, of the individual plants which are in this abnormal state, have been affected in some strange manner by the conditions to which they themselves or their parents have been exposed; but whilst thus rendered self-sterile, they have retained the capacity common to most species of partially fertilizing and being partially fertilized by allied forms. However this may be, the subject, to a certain extent, is related to our general conclusion that good is derived from the act of crossing. Gaertner experimented on two plants of _Lobelia fulgens_, brought from separate places, and found[308] that their pollen was good, for he fertilised with it _L. cardinalis_ and _syphilitica_; their ovules were likewise good, for they were fertilised by the pollen of these same two species; but these two plants of _L. fulgens_ could not be fertilised by their own pollen, as can generally be effected with perfect ease with this species. Again, the pollen of a plant of _Verbascum nigrum_ grown in a pot was found by Gaertner[309] capable of fertilising _V. lychnitis_ and _V. Austriacum_; the ovules could be fertilised by the pollen of _V. thapsus_; but the flowers could not be fertilised by their own pollen. Koelreuter, also,[310] gives the case of three {137} garden plants of _Verbascum phoeniceum_, which bore during two years many flowers; these he successfully fertilised by the pollen of no less than four distinct species, but they produced not a seed with their own apparently good pollen; subsequently these same plants, and others raised from seed, assumed a strangely fluctuating condition, being temporarily sterile on the male or female side, or on both sides, and sometimes fertile on both sides; but two of the plants were perfectly fertile throughout the summer.