Thames plants must strike every one as belonging to an ancient order of life. But the vast clouds of winged ephemeridae that dance over its waters when there is a rise of "May-fly" in early summer look to be not only the creatures of a day, but of our day. In the astonishing wave and rush of life seen at such times, when from every plant and pool winged creatures are ascending to float in air, it is difficult to picture the silence and stillness of a world where there were no birds, or hum of bees, and no signs of the other insects which exceed the other population of the earth by unnumbered myriads of millions; yet the insects, even the same identical species which dance over the Thames to-day, are among the very oldest of living things, just as its plants and its shells are. Rocks and slate are not ideal butterfly cases; and if the fragile limbs of the beetle and grasshopper of the successive prehistoric worlds had perished beyond the power of identification, no one could have felt surprise. But such has been the industry of modern naturalists - to give the widest name to those who have devoted their time to the search for, and description of, fossil insects - that the remains of thousands of species have been identified, and the time of their appearance upon the earth approximately fixed. The latest contributor to this elegant branch of the study of fossils is Mr. Herbert Goss.[1] Perhaps the most interesting of his conclusions is the antiquity, not only of the existing orders of insects, but even of their particular families and genera, as compared with vertebrate animals. It is astonishing to find not only crickets and beetles existing at periods enormously earlier than the appearance of birds or fish, but that they conformed in type to the families in which they are classed to-day. Though they become fewer and fewer as they are tracked back up the river of time, there are not found in the earliest fossil-bearing rocks any connecting links or earlier and simpler forms of insect life, or a clue to the common ancestor of insects, spiders, and shrimps, which naturalists would dearly like to discover. There is a baffling completeness about these creatures. When in the lias period, for instance, the vertebrates were huge saurian reptiles and flying lizards, and scarcely any of our existing classes of fish had come into existence, the beetles, cockroaches, crickets, and white ants were there, with all the distinguishing characteristics of the existing families as they were settled by Linnaeus.

The first insect known to have existed, a creature of such vast antiquity that it deserves all the respect which the parvenu man can summon and offer to it, was - a cockroach. This, the father of all black-beetles, probably walked the earth in solitary magnificence when not only kitchens, but even kitchen-middens were undreamt of, possibly millions of years before Neolithic man had even a back cave to offer with the remains of last night's supper for the cockroach of the period to enjoy. His discovery established the fact that in the Silurian period there were insects, though, as the only piece of his remains found was a wing, there has been room for dispute as to the exact species. Mr. Goss in his preface to the second edition of his book notes that what is probably a still older insect has been found in the lower Silurian in Sweden. This was not a cockroach, but apparently something worse. If the Latin name, Protocimex Silurius, be literally translated, it means the original Silurian bug. It was a fair conjecture that insects appeared about the same time as land plants first grew on the earth. As almost all the species either feed on some vegetable substances in growth or decay, or else live upon other insects, some such provision of food was necessary for them. Remains of such plants were discovered in the Silurian rocks. In the Devonian formations, which contain the next oldest set of fossil insects, numbers of conifers and ferns are found. Yet even then the only vertebrate animals seem to have been fish. The insects still had the land all to themselves. Of one of these Devonian insects the base of a wing was the only part preserved in the rock. From this it was possible to tell the order to which the creature belonged. It was one of the Neuroptera - insects with wings in which the veins run straight down the wing, sometimes joined by cross branches at right angles. Some of the modern kinds are very beautiful four-winged flies, with bright colours on their wings like butterflies. Others are ant-lions or caddis-flies. The curve of the fragment of wing also suggested its probable size when unbroken. It was perhaps two inches long. As there are little horny rings round the wing base like those which crickets have, on which they rub their legs and so "chirp," it is also quite likely that this insect of hoary antiquity did the same, and enlivened the silence of Devonian fern groves with a prehistoric hum. It is quite in keeping with modern ideas that in that age of fishes one of the most remarkable insects should have been a kind of May-fly, "a large species of Ephemerina, which must have measured five inches in expanse of wings." Thus our Thames May-flies had gigantic prehistoric ancestors, which appeared on earth, possibly with their present associates the caddis flies, at an enormously remote age.