Reviews — The Reptiles of the Karroo Forination. 519 



obviously due to a drop in the summer temperature. A drop in the 

 mean annual temperature cannot be deduced, but is probable. The 

 important point is, however, that the halt is due to a temperature 

 change, and not to a change in the precipitation or nourishment of 

 the ice-sheet. This conclusion, which was also reached by De Geer 

 in Sweden, is in itself very interesting, as it is obvious that a halt 

 might be conceived to result from increased precipitation on the 

 snowfields, producing an active thrust forward of the glacier front, 

 and so balancing the marginal wastage. This, however, is only 

 incidental to the main point, which it is the object of this review to 

 emphasize, and which appears to be quite novel. 



When the successive positions of the ice-margin are laid down on 

 a map, or, more strikingly still, when the rate of retreat is plotted 

 as a curve against the years, it becomes apparent that periods of 

 halt in retreat were followed in nearly all cases by unusually rapid 

 withdrawal, which gradually in the course of ten or fifteen years 

 reverted to the normal rate characteristic of that stage of the retreat 

 in general. This does not appear to have been due to excessive 

 ablation, for the summer sediments in the laminated clays corre- 

 sponding to these periods of rapid retreat have no exceptional 

 thickness, and Dr. Sauramo draws the conclusion that during the 

 cold period of the halt the ice-sheet had become thinner and so 

 yielded more rapidly at its margin, once melting set in again. 



Here, then, is additional light on the climatic condition of the 

 halt-stages. Wastage was in progress during these periods, but was 

 not, to any great extent, effected by the process of ablation. The 

 ice surface was being lowered by some agent independent of ablation. 

 There seems to be only one physical process that could account for 

 such a remarkable phenomenon, namely, direct evaporation from the 

 ice. This is, of course, well known to be physically possible, but it 

 has certainly never been appreciated by geologists as an effective 

 agency in glacial control. If it be now admitted as a result of this 

 work of Dr. Sauramo's that such is the case, and that under 

 sufficiently low conditions of temperature the wasting effects of 

 direct evaporation may become comparable with those of ablation, or 

 possibly exceed them, such a conclusion must modify all our notions 

 of the growth and ultimate decay of ice-sheets. 



W. B. Wright. 



Contributions to the Knowledge of the Reptiles of the 

 Karroo Formation. By Dr. E. C. N. van Hoepen. Annals 

 of the Transvaal Museum, vol. vii, pt. ii, 1920. 



T^HE author describes a new genus of Dinosaur from the Stormberg 

 -*- beds, and designates it Aristosaurus erectus. The animal was of 

 small size and highly specialized. It shoM^s adaptation to a bipedal 

 mode of locomotion. The humerus is relatively much shorter than 

 in other Anchisauridae, the family to which it belongs. 



