THERE is now not the slightest doubt that the majority of the commonest fern-like fronds of the Coal Measures did not belong to the ferns at all, but were-plants of an entirely extinct group. In several cases the have actually been found borne on the fronds, sometimes on a naked part, but sometimes on fronds with normal leaflets. There was nothing in the nature of a specialised flower.
In appearance many of the-ferns must have resembled the tree-ferns of the present day. In the structure of the as well as of the reproductive organs they show an amazing variety, and in spite of the superficial resemblance it does not seem that these plants were at all closely related to any known ferns, either fossil or recent. Moreover, the earliest-known forms of the ferns and of the seed-ferns found in Upper Devonian rocks are not in the least alike. Both may have had a common origin, but if so it must have been in an earlier era. The seed-ferns may be related to the gym-nosperms, especially the cycads, for there are similarities in the anatomical characters of the as well as in the structure of the seed.
In the fertilisation of the seed-ferns we have an interesting example of the dependence of highly-developed land-plants on water at one stage of their life-history. Fertilisation in the ferns and other spore-plants, and also in the cycads, still takes place by the liberation of an active male cell which requires the presence of water in order to swim to the female organ. The higher seed-plants of to-day have got beyond this dependence on water, and the male cell is passive, but theof the Carboniferous seed-ferns resemble those of cycads in having a pollen-chamber into which when it was filled with water the male cells were no doubt liberated. In some of these fossil seeds pollen-grains have actually been found in the pollen-chamber.
Palaeozoic seeds also differed from those of modernin another important respect: they did not contain an embryo. Some of them were quite small—only a few millimetres in length—but a few were as large as a duck’s egg, with a very thick coat. The question of whether the
higher seed-plants are all descended from a single stock is still undecided. Many botanists, however, are inclined to see in the network of forms grouped together as the seed-ferns the starting-point of the several lines of gymnosperms, and also of the flowering plants of to-day. This problem will be referred to again below.