November 22, 191 7] 



NATURE 



23. 



obtained on 333 days; when possible, duplicates were 

 taken at short intervals for the detection of spots of 

 brief duration. In addition to the usual system of tele- 

 graphic time-signals, arrangements have been com- 

 pleted for the daily transmission of a wireless signal 

 for the use of shipping in South African waters. 



THE CLASSIFICATION OF THE 



BRITTLE-STARS.^ 



'T'HE Ophiuroidea have long presented a problem 



•^ to the systematist, and its solution was not ad- 

 vanced when the palaiontologist joined the neontologist 

 in council. The reason is twofold : the modern repre- 

 sentatives of this Echinoderm class differ little in 

 great points, but greatly in littl^oints; the Palaeozoic 

 representatives, which do differ much, and should 

 throw light on the origins of orders, are so preserved 

 as to be difficult of interpretation. Twenty-hve years 

 ago Mr. Jeffrey Bell divided the recent forms accord- 

 ing as they could only move the arms horizontally or 

 could also coil them vertically, the latter being re- 

 garded as more primitive. Dr. J. W. Gregory ex- 

 tended this system by adding an order for those yet 

 more primitive forms in which the arm-bones still con- 

 sisted of the origmal paired elements. It was early 

 pointed out that these divisions represented successive 

 grades rather than divergent orders; but doubt has 

 since been cast even on their correspondence with 

 reality by the observations of Schondorf, Sollas, Mor- 

 tensen, and Spencer on the older fossils and on the 

 crucial genus Ophioteresis. Now a voice from the 

 East complains : " 1 found the classifications very un- 

 satisfactory. Indeed, their imperfections became a haunt 

 to me." From a study of recent genera, Mr. Matsu- 

 n-oto infers that in respect to both mouth-frame and 

 arrn-bones the forms which can only move their arms 

 horizontally are more primitive than those which can 

 coil them vertically. He therefore rejects any system 

 based mainly on the joint-faces, and puts forward a 

 classification of his own. 



The difficulty presented by the Palaeozoic forms is 

 evaded by separating them as a sub-class : GEgophiu- 

 roida. ^ Since this admits no genera with ventral arm- 

 plates it cannot quite correspond with the Palophiurse 

 (Haeckel), but its difference from the Protophiuroidea 

 (Sollas) is not obvious. Neither is it clear whether the 

 author would regard the CEgophiuroida as a non- 

 jpersistent group parallel to both Asteroidea and Ophiu- 

 roidea, or whether he would bring it into the ancestry 

 of modern brittle-stars. 



All normal Ophiuroidea with the ventral surface of 

 the arms covered by plates are constituted a sub-class 

 Myophiuroida. Its Palseozoic representatives have no 

 distinct plates in the skin of the central disc, the 

 mouth-frames are slender, dorsal arm-plates are absent 

 or mcipient, ventral arm-plates are small and depressed 

 below the projecting edges of the side plates. Among 

 recent forms it is the Ophiomyxinae that come nearest 

 to this condition, but it is afso approached by those 

 Ophiacanthidae in which the arms are only flexible 

 horizontally. From the Ophiomyxidae Mr. Matsumoto 

 derives all the Trichasteridae and GorgonocephalidjE, 

 and separates the three families as an order Phryno- 

 phiunda. 



From the early Ophiacanthidae are supposed to spring 

 all the other Ophiuroids, diverging along three lines. 

 I he first of these passes, through those Ophiacanthidee 

 which can coil the arms vertically, to the Hemieury- 

 alidae ; and these two families compose the order Leemo- 

 phiurida. The two other lines never attain vertical 



rLlit3Z°^''^^ °U*P""«« (Jphiuroidea, arranged according to a Ne«r 

 Classification By Hikoshichiro M.-,tsumoto. Journ. Coll. Scie,rce, 'lokyo 

 vol. xxxvui, Article 2. Pp. 408 + vii plates. (Un--'- " • '-5.''?''Jo, 

 1Q17.) 



NO. 2508, VOL. 



coiling. From one another they are distinguished in 

 the articulation of the radial shield and genital plate : 

 in the one case this is by a single ball-and-socket joint, 

 in the other case by two condyles and sockets. The 

 former line passes, through the' Amphilepididae, to the 

 Amphiuridae and Ophiotnchidae, and, since these two 

 families have stout mouth-frames and teeth, the whole 

 order is called Gnathophiurida. Along the other line 

 arises a host of forms, divergent in structure and com- 

 plex in relationship, which are grouped under five 

 families : Ophiodermatidas, Ophiochitonidae, Ophio- 

 comidse, Ophiolepididae, .ind Ophioleucidae. 



Mr. Matsumoto's classification, being essentially 

 phvlogenetic, will have to be checked by the palaeonto- 

 logist before it can be considered established. The 

 morphological bases, however, seem well selected and 

 are well illustrated. F. A. B. 



Pp. 408 + vii plates. (University, Tokyo, March' 



100] 



ATMOSPHERIC POLLUTION. 

 'pHE second report (1915-16) of the Committee on 

 ■*■ Atmospheric Pollution has just been issued in 

 the form of a supplement to the Lancet, the delay in 

 its appearance being due to lack of funds. This diffi- 

 culty has now been met by the receipt from the 

 Department of Scientific and Industrial Research of a 

 Government grant, which provides the necessary 

 equipment for collecting and analysing the smoke de- 

 posits at different centres. The work, moreover, has 

 been given official approval and status by placing it 

 under the control of the Meteorological Office, the 

 committee being constituted as an advisory committee 

 of that department. 



Owing to the dejpletion of the staffs formerly 

 collaborating in these investigations, certain stations 

 have found it impossible to continue observations, so 

 that the list for the year is curtailed. The general 

 methods of analysis and tabulation of results are, with 

 slight modifications, those previously described in the 

 former report (Nature, May 4, 19 16, vol. xcvii., 

 p. 203). 



It is interesting to note from the point of view of 

 fuel economy that the deposit for the year in the 

 County of London alone, which consisted of matter 

 derived from waste fuel in the form of smoke, amounted 

 to 54,200 tons. The report adds that not only is it 

 necessary to scrutinise carefully every source of "waste, 

 but it is equally necessary to conserve the health and 

 physical energy of the people. From this point of view 

 it refers to the fact that the average weight of air 

 consumed per day by the adult human being is 30 lb., 

 as compared with 7-2 lb. of solid food and water. 



In the section devoted to a discussion of results a 

 comparison is drawn between the total solids deposited 

 in the six summer and the six winter months of 1915-16 

 with the corresponding periods of 1914-15. Without 

 reproducing details of the results it may suffice to say 

 that in the larger number of centres there has been 

 an increase in the amount, a few centres in the Man- 

 chester and Glasgow area showing a decrease in the 

 winter months, while Birmingham Central, Bolton, 

 Malvern, Sheffield, and York show a diminution in the 

 summer months. Of the actual quantities, the mean 

 monthly deposit in tons per square kilometre is tabu- 

 lated for the different centres. It appears from this 

 that Oldham has the distinction of showing the largest 

 deposit of total solids, carbonaceous matter other than 

 tar, and insoluble ash, while Glasgow occupies the 

 highest place in ammonia, sulphates, and tar. Mal- 

 vern shows the minimum deposits in nearly every item. 

 If there were the same fierce rivalry between towns 

 as existed in medieval Italy, we might hope that 

 industrial centres might vie with Malvern in improv- 

 ing their atmosphere. 

 There seems very little prospect of any such peaceful 



