January 3, 1901] 



NATURE 



233 



suitable screens between the source of light and the ob- 

 ject to be illuminated ; but it appears doubtful whether 

 enough light can be easily obtained in this way for very 

 high magnifications. 



With regard to the selection of the specimens to be 

 examined, it is well known that much time is wasted 

 when working out a complete series of alloys of two 

 metals. It is necessary to prepare, polish and etch a 

 series of specimens, many of which will present no fea- 

 tures of interest when examined under the microscope. 

 M. Le Chatelier proposes to shorten the search for 

 typical alloys by melting together two superimposed 

 layers, each consisting of a pure metal, the lighter one 

 being on the top. If no alloys are formed of greater 

 density than the heavier metal, and the crucible is al- 

 lowed to cool undisturbed, a culot can be obtained which, 

 on being sawn through vertically, shows a complete grada- 

 tion from one pure metal to the other, passing through 

 the whole series of alloys, which can then be studied in 

 one specimen. In this way he prepared a number of 

 series, the three illustrations (Figs, i, 2 and 3) being from 



Fig. 3.— Crystals of AICU3. 



photographs of different parts of a single specimen show- 

 ing the aluminium-copper series. Fig. i shows crystals 

 of AljCu ; Fig. 2, crystals of a compound of undeter- 

 mined composition which is not far from that expressed 

 by the formula AlCu ; Fig. 3 shows crystals of the com- 

 pound AICU3. It would seem that the exact percentage 

 of any particular part of a specimen prepared in this 

 way must be a matter of uncertainty, but there is no 

 doubt that, in the hands of M. Le Chatelier, the method 

 has already yielded some interesting results. 



T. K. Rose. 



SOME RECENT ADVANCES IN GENERAL 



GEOLOGY. 

 A MONG the recent researches on organic remains 

 ■^*- none are of greater geological interest and import- 

 ance than those relating to the Radiolaria. The tiny 

 siliceous structures which belong to this Order of 

 Protozoa have long been recognised in our formations, 

 but the part they have played in building up portions 

 of the stony structure of the earth has not until lately 



NO. 1627, VOL. 63] 



been realised. The most striking evidence was that 

 brought forward by Prof. Edgeworth David and Mr. 

 E. F. Pittman {Quart. Journ. Geol. Soc, vol. Iv. p. 16, 

 1899). They describe a great series of siliceous lime- 

 stones, jaspers and claystones, with interstratified coral 

 limestones and plant-beds, and submarine tuffs, the 

 whole attaining a thickness of over 9000 feet, and ex- 

 tending over many hundred square miles in New South 

 Wales. In the bulk of these rocks Radiolaria are present 

 at the rate of about one million to the cubic inch, and 

 among the forms Dr. G. J. Hinde has recognised 

 twenty-nine genera and fifty-three species. Taken 

 as a whole, the deposits are fine-grained, and bear 

 evidence of having been laid down in clear sea- 

 water, beyond reach of any but the finest sediment. 

 They do not indicate any very considerable depth ot 

 water ; but they tell of a vast lapse of time, and of con- 

 ditions which prevented the dispersal over the area of 

 coarse detritus. What exactly were these conditions it 

 remains for future research to discover. In this country, 

 in Devonshire and Cornwall, the occurrence of radiolarian 

 cherts, both of Ordovician and Carboniferous ages, has 

 been made known through the observations of Dr. Hinde, 

 Mr. Howard Fox and Mr. Teall. The more prominent 

 of these rocks are found in the Lower Carboniferous 

 formation of Coddon Hill near Barnstaple, where the 

 chert-beds have long been known, although their organic 

 origin was not until recently discovered. The freedom 

 of the beds from mechanically-formed detritus has led to 

 the supposition that these strata were deposited in deep 

 water and at some distance from the coast, although the 

 associated strata above and below the chert-beds do not 

 lend support to the hypothesis. The fact is that at the 

 present time the only extensive radiolarian deposits 

 known to be in process of accumulation are in the deeper 

 oceanic regions. 



Radiolaria, while entirely marine, are widely distri- 

 buted, and they can exist at various depths in deep and 

 shallow seas. It may be surmised, therefore, that in 

 shallower areas coral-reefs may have acted as barriers to 

 the dispersal of terrestrial debris. Hence in our explan- 

 ations of the physical conditions of the past we must be 

 guided by the general characters of the sedimentary 

 strata in which bands and beds of radiolarian chert occur, 

 rather than by the evidence of the chert itself. There is, 

 however, little doubt, from the wide distribution of these 

 lowly forms of life, that they may prove of considerable 

 importance in the identification of horizons, although, as 

 might be expected from their present geographical and 

 bathymetrical ranges, some specific types have been of 

 long geological duration. 



In the coast ranges in California, and again in Borneo, 

 such radiolarian rocks of Jurassic, or possibly Lower Cre- 

 taceous age occur, and it is noteworthy that Dr. Riist has 

 remarked that " the differences in the Radiolaria from 

 these two rock-divisions are not very striking." (See 

 Hinde's " Description of Fossil Radiolaria from Central 

 Borneo," 1899.) 



The question whether the Wealden strata which are 

 essentially freshwater should be grouped as Jurassic 

 rather than Cretaceous has been raised by geologists in 

 the New as well as in the Old World, who have argued 

 that the Wealden plants, fishes and reptiles are Jurassic 

 rather than Cretaceous in character. There has never 

 been any question in this country that the Purbeck and 

 Wealden Beds are intimately connected both strati- 

 graphically and palceontologically, and it has been held 

 by some geologists that locally the Wealden Beds and 

 Lower Greensand bear also evidence of continuous 

 deposition. The subject was lately discussed by Mr. 

 G. W. Lamplugh (Brit. Assoc, Bradford, 1900), who 

 points out that in Dorset, Hampshire and Surrey there 

 is evidence of the close stratigraphical connection be- 

 tween Wealden and Lower Greensand, that part of the 



