2 6 PROCEEDINGS OF THE 



these organs were not yet quite sharply differentiated as roots ; 

 at any rate there are certain points in which they I'ather resemble 

 modified leaves, though in my opinion the root-characters pre- 

 dominate. Though they may thus be " primitive," from the 

 point of view of our current morphological categories, these 

 organs, as Prof. Weiss has discovered, show a remarkable adaptive 

 mechanism in the presence of strands of water-conducting 

 elements, running out from the central vascular bundle and 

 terminating in plates of tracheae placed in the outer cortex. The 

 whole constitutes an absorptive apparatus more elaborate than 

 anything found in recent roots, if we except a few highly 

 specialized haustorial roots of parasites. This example seems to 

 me instructive, for it shows how a very high degree of adaptation 

 may coexist with characters which suggest a somewhat archaic 

 type of organ. 



As an example of adaptation to more special conditions, I may 

 instance the xerophytic characters shown by various Carboniferous 

 plants, especially in the structure of their leaves. In the 

 Lepidodendreae, the stomata were commonly restricted to two 

 deep furrows on the lower side of the leaf, where they were 

 further sheltered by a growth of hairs. The whole structure 

 of the leaf suggests a xerophilous habit. The late M. Renault 

 regarded the transversely elongated cells of the mesophyll in 

 SigiUaria as a means of rolling up the leaf, to diminish ti'an- 

 spiration, as occurs, for example, in certain grasses at the present 

 time. In the Pteridosperm Lyghiodcndron the leaflets of the 

 fern-like fronds had a fleshy character, and a conchoid, incurved 

 form ; they were provided with a hypoderma, and the endings of 

 the vascular bundles were often dilated, perhaps in connection 

 with glands. These are all characters such as are met with in 

 the plants of salt-water swamps at the present day. 



The subject of the physiological anatomy of Palaeozoic plants 

 has never yet been attacked in a systematic manner. In a 

 conversation I had with Prof. Haberlandt of Graz, four years ago, 

 he said that he would like to undertake their investigation from 

 this point of view ; if he would do so it is certain that a remark- 

 able advance in our conceptions of the adaptive structure of ancient 

 forms would result. Even with our present limited knowledge, 

 however, it is sufficiently clear that the plants of that relatively 

 (but only relatively) early period were thoroughly well adapted to 

 the conditions of their life ; succeeding ages bi'ought with them 

 neiv rather than hetter adaptations. 



Though there is no question of absolute perfection in Nature, 

 it appears that under given conditions, adaptation is and was 

 sufficiently perfect to make it very difficult to put one's finger on 

 any defect. AVhen we think we can do so it generally turns out 

 that the defect is in the mind of the critic rather than in the 

 organism criticised. AVe will take a particular case, where the 

 history seems to give some justification for our fault-finding. 



The late Palaeozoic family Medulloseae were in some respects 

 the most i-emarkable plants from an anatomical point o£ view that 



