October 12, 191 ij 



NATURE 



503 



Reid asks, why do we not also find in Ireland 

 Portugal laurels and oaks, which are hardier than 

 Arbutus? An alternative view is that Arbutus and its 

 compatriots came to Ireland after the Glacial period, not 

 by migration overland, but by natural means of dispersal, 

 which would be favoured by the small size of their seeds 

 as contrasted with the larger seeds of oaks and other 

 southern plants, which never reached our shores. 



The possibilities of plant dispersal by natural agencies 

 have recently received a striking demonstration in the 

 recolonisation of the island of Krakatau, in the Sunda 

 Straits. In 1883 Krakatau, then covered with a dense 

 tropical vegetation, was partially destroyed by a series of 

 exceptionally violent volcanic explosions. It is believed 

 that no vestige of life remained. In 1906, twenty-three 

 years after the sterilisation of the island, 137 species of 

 plants were collected, and the vegetation was in places so 

 dense that a party of botanists penetrated with the greatest 

 difficulty beyond the coastal belt. Some of the trees had 

 reached a height of 50 feet. The nearest islands, except 

 the small island of Sebesi, about twelve miles distant, are 

 Java and Sumatra, separated from Krakatau by a stretch 

 of water twenty-five miles in breadth. This new flora, 

 introduced by ocean currents, by wind, and by the agency 

 of birds, affords a useful object-lesson in regard to the 

 efficiency of plant-dispersal without the aid of land- 

 connections. 



Before passing to the consideration of questions necessi- 

 tating frequent reference to different periods of geological 

 history, it is essential to direct attention to the sequence 

 of chapters as revealed by the earth's crust. It is from 

 the scanty records of plant life preserved in the sedi- 

 mentary deposits of former ages that we obtain such 

 evidence as we have in regard to the relative antiquity 

 of different types. The crust of the earth, as Darwin 

 wrote, " with its imbedded remains, must not be looked 

 at as a well-filled museum, but as a poor collection made 

 at hazard and at rare intervals." 



The oldest rocks, largely composed of gneisses and other 

 products of igneous action, throw no light on the nature 

 of the organisms which existed in the earliest epochs of 

 the earth's history. The complex structure of the oldest 

 known animals and plants compels us to believe that they 

 are the descendants of simpler forms of much greater 

 antiquity. As the Cambridge professor of modern history 

 has aptly said, " All the epochs of the Past are only a 

 few of the front carriages, and probably the least 

 wonderful, in the van of an interminable procession." 

 The foundation-stones of the earth's crust have been so 

 much folded and altered in the course of geological time 

 that it is no wonder they have been searched in vain for 

 records of primitive life. Passing higher up the series 

 through the vast thickness of Cambrian, Ordovician, and 

 Silurian strata, 1t is in the Devonian rocks of Ireland and 

 elsewhere that we first discover the records of Palaeozoic 

 floras, and these bear a fairly close resemblance to the 

 still richer floras of the succeeding Carboniferous epoch. 



Turning, now, to the top of the geological series, the 

 submerged forests round our coasts and the tree-stumps 

 buried in peat form connecting links between existing 

 plants and those of a prehistoric age. A little further 

 down occur the Boulder Clay and other legacies from the 

 Glacial period, and below these are the fragmentary relics 

 of a pre-glacial vegetation. Further down the plants 

 become less familiar and show a closer agreement with 

 those of subtropical and tropical countries than with the 

 recent vegetation of Britain. From the London Clay, a 

 marine deposit, which underlies London and Portsmouth, 

 and is exposed in the Isle of Wight, the Isle of Sheppey, 

 and in other places, fossil seeds and fruits have been found 

 practically identical with those of existing tropical species. 

 One of the London Clay fruits may be mentioned as an 

 especially interesting sample of the early Tertiary flora, 

 namely, the genus Nipadites, so called because of the verv 

 close resemblance it bears to the fruits of the common 

 tropical plant Nipa. Nipa fruticans, often described as a 

 stemless palm, grows in the brackish estuaries of manv 

 tropical countries. The occurrence of fossil fruits of this 

 type in Tertiary beds in England, Belgium, and France 

 affords a striking instance of changes in the distribution 

 of an ancient plant now restrictpd to warmer regions. 

 NO. 2l8g, VOL. 87] 



Below the Tertiary rocks we descend to the Chalk 

 period, when a clear and comparatively deep sea covered 

 the areas now occupied by chalk downs and cliffs. 

 " During the Chalk period," as Huxley wrote, " not one 

 of the present great physical features of the globe was ir. 

 existence. Our great mountain ranges, Pyrenees, Alps, 

 Himalayas, Andes, have all been upheaved since the 

 Chalk was deposited." Below the Chalk in the Weald 

 district of Kent and Sussex a rich Wealden flora has been 

 discovered in sediments laid down in a shallow lake which 

 occupied the south of England and extended across what 

 is 'now the English Channel. The Wealden plant-beds 

 have as yet furnished no satisfactory specimen of a flower- 

 ing plant. It is one of the most interesting facts in the 

 history of the vegetable kingdom that the highest class 

 of plants which now overspreads almost the whole world 

 did not come into prominence until after the close of the 

 Wealden period. When, in the course of evolution, the 

 flowering plant became a competitor in the struggle for 

 existence, it spread with amazing rapidity. 



The sandstones and shales in the Yorkshire cliffs and 

 slightly older rocks on the Dorsetshire coast have supplied 

 data which enable us to reconstruct in some measure the 

 vegetation of that stage in the earth's history known as 

 the Jurassic period. Below the Jurassic strata in some 

 parts of the world, as in the south of Sweden, Germany, 

 and elsewhere, from rocks of Rhaetic age numerous fossil 

 plants have been obtained. Descending further, the plant 

 records from strata belonging to the early days of the 

 Triassic period afford evidence of changes in the nature 

 of the vegetation, which become more pronounced in the 

 still older Permian and Carboniferous floras. It is, how- 

 ever, not my intention to deal with the vegetation of the 

 coal forests and other Palaeozoic floras ; they are com- 

 posed of plants for the most part much less closely related 

 to existing types than those which I have selected as 

 examples of links with the past. 



Probably the human race made its entrance on to the 

 world's stage at some time during the Tertiary period ; but 

 in Britain abundant evidence of man's presence is not met 

 with until after the Glacial period. An additional illus- 

 tration of the enormous antiquity of some of the plants 

 to be described later is furnished by the absence of flower- 

 ing plants in the rich floras below, or even including, those 

 of Wealden age. 



The widely distributed class of ferns supplies some 

 notable instances of links with the past. It is not in- 

 frequently the case that plants which are now characterised 

 by a restricted geographical range have a wide range in 

 time, and, conversely, plants which are now more or less 

 cosmopolitan may be of comparatively recent origin. 

 Antiquity and restricted distribution often go together. 

 The bracken fern, which we are apt to regard as essenti- 

 ally British, occurs also in Tasmania, in the Malay 

 Peninsula, in British East Africa, in the Himalayas, and 

 in many other countries ; it is one of the most cosmopolitan 

 of all living ferns. It would seem probable that this 

 injurious species is one of the more modern members of 

 its class. On the other hand, the family to which the 

 royal fern (Osmunda regalis) belongs, though widely spread 

 at the present day, has been traced into the Palaeozoic 

 era. The recent researches of Dr. Kidston and Prof. 

 Gwynne Vaughan have demonstrated the existence in the 

 Permian flora of Russia of ferns exhibiting a close rela- 

 tionship to existing members of the Osmunda family. The 

 section shown on the screen was cut from a petrified fern 

 stem of Permian age, which shows in its anatomical 

 characters a remarkable resemblance to a stem of 

 Osmunda. Some of the commonest fern leaves in the 

 Jurassic rocks of Yorkshire and in many other parts of 

 the world, both north and south of the equator, may be 

 referred with a considerable degree of certainty to the 

 Osmunda family, the relationship being indicated, not by 

 mere external resemblance, but by the structure of the 

 spore-capsules, and in a few cases by the occurrence of 

 petrified stems. It is by no means improbable that the 

 royal fern and other existing members of the family are 

 entitled to the distinction of an ancestry which extends 

 farther back into geological time than that of any other 

 section of living ferns. 



The next example of an old type of fern is one which 



