Jtme 2 1, 1877] 



NATURE 



139 



spinulosa, and have concluded that whilst their Diploxy- 

 loid organisation differs from that of the Lepidodendra, 

 it justifies the conclusion that Sigillaris were not lycopo- 

 diaceous but gymnospermous plants. But I have already 

 shown that several indisputable Lepidodendra have pre. 

 cisely the same organisation. Hence I contend that 

 Brongniart's reasons for separating these plants have no 

 existence, and consequently his conclusions must be 

 abandoned ; M. Grand' Eury, forgetting this part of my 

 work, and only remembering that I have also described 

 the bark of a true Syringodendroid Sigillaria, and shown 

 that it is identical in every feature with the corresponding 

 tissue in Lepidodendra, says that I have arrived at my 

 conclusion " par des faits isoles d'apres I'analogue de 

 Tecorce, et non par des exemples complets reunissant les 

 caracteres extdrieurs aux caracteres intdrieurs." How, in 

 the face of my published memoirs, my friend could make 

 so erroneous a statement, I am at a loss to conceive. 



I should have felt it necessary to have subjected the 

 volumes under consideration to an incisive criticism on 

 these and some similar points, were it not that he kindly 

 allows me to quote from some letters which I have received 

 from him. In these communications he says : — " Les 

 points sur lesquels nous difierons sont prccise'ment ceux 

 que je n'ai pas etudics." Referring to facts which I have 

 observed, he adds ; " Comme je n'avais pas ces faits pour 

 me guider, j'ai conclu d'apres ce que je connaissais bien 

 et je ne suis pas fain de conclure que je me suis trompe ; 

 dans ce cas nous aurions dans les Sigillaires et les Lepi- 

 dodendrons des cryptogames excessivement elevees en 

 organisation ; si elevees qu'ils formaient, en quelque 

 fa(;on, une classe intermediare entre ces plantes et les 

 Gymnospermcs." " L'Association presque constante des 

 macrospores avec le debris des Sigillaires est en faveur 

 de vos conclusions." Quite in accordance with the above 

 remarks are the following observations which the author 

 makes in his volumes : " II est au moins curieux que, a 

 part le corps vasculaire, les autres parties des Sigillaires 

 soient semblables aux parties correspondantes des Lepi- 

 dodendrons." This is perfectly true with the exception 

 that the vascular portions are less exceptional than M. 

 Grand' Eury's remarks imply. The Sigillarian stem is 

 merely that of Lepidodendron Harcourtii, with an addi- 

 tional exogenous vascular zone interposed between that 

 of the Lepidodendron and its investing cortex ; and 

 ■which I Jind in otlur true Lepidodendra. If all the 

 plants of the coal-measures which possess a similar exo- 

 genous zone are to be transferred from the cryptogamic 

 to the phanerogamic group, there will be few cryptogams 

 left in the carboniferous rocks beyond Lepidodendron 

 Harcourtii and the ferns. M. Grand' Eury concludes 

 his notice of the Sigillariae by a remark which I fully 

 endorse : " Seulement je crains d'avoir tire des conse- 

 quences trop completes de ces indices insuffisants, dont 

 je n'aurais peut-etre alorsmeme du parlerque pour eveiller 

 I'attention des obsei-vateurs sur une solution possible du 

 plus important probleme de la paleontologie vegctale." 



Having thus indicated some very important points 

 respecting which I am compelled to differ from M. 

 Grand' Eury, I can with the sincerest truth again 

 express my sense of the value of this new contri- 

 bution to the study of the carboniferous flora, and 

 of the praiseworthy perseverance with which the author 



has laboured for many years in collecting his materials. 



The most prominent fact which the work reveals is the 

 remarkable abundance of the Cordaites in the coal- 

 measures of Central France, compared with what we see 

 in England. In some districts, as M. Grand' Eury 

 informs us, the coal is almost entirely composed of their 

 debris. I have met with nothing like this in Great 

 Britain, but it is in strict accordance with what we know 

 of the distribution of living plants, that whilst similar 

 types may be expected to be met with over wide geo- 

 graphical areas, some forms will predominate in one 

 region, whilst in other localities different types will 

 prevail ; hence the materials out of which coal has been 

 found must have been widely different at these various 

 spots. 



The plates with which the above work is illustrated are 

 extremely beautiful, as is usually the case with the pro- 

 ductions of the French lithographers. 



W. C. Williamson 



OUR BOOK SHELF 

 Annces do Observatorio do Infante D. Luiz. Magnetisnio 



Terrestre. Lisboa, 1876. 

 This part of the Annals of the Lisbon Observatory is a 

 continuation of those noticed in Nature, vol. xiii. p. 301. 

 The results for the magnetic declination are carried 

 forward from 1867 to 1S71, while some include the means 

 from 1S58 to 1875. This is the case for the secular 

 change and annual variation. Mr. Capello found pre- 

 viously that the north end of the declination magnet 

 approached the north at the rate of 5''9i yearly (1858- 

 1868). The results he now divides into two series, 1858 to 

 1S66, with a rate of 5 ''46, and 1866 to 1875, with a rate of 



7 ''64 yearly. 



The yearly means are deduced from observations at 



8 A.M. and 2 p.m. Mr. Capello has also shown that the 

 diurnal law of disturbance appears to be different at 

 Lisbon in different years of the decennial period.' In this 

 case, even if two observations daily were otheiwise suffi- 

 cient to give accurate means, or means strictly compa- 

 rable from year to year, the varying effect of the disturb- 

 ance on the observations at the two hours mentioned 

 would of itself interfere with this comparability. It is 

 probably for these reasons that the yearly means at 

 Lisbon do not appear to show the small decennial in- 

 equality in the secular movement first indicated by me 

 in 1857, and afterwards discovered by Hansteen and 

 Lloyd. 



Mr. Capello has repeated discussions for the magnetic 

 disturbances with the increased materials in his posses- 

 sion. He had observed in a preceding number of the 

 Annals, that many observations which were considered 

 disturbed (that is to say, which differed from the means 

 for the hours by 2''26 or more) really belonged to diurnal 

 variations which were regular, only larger than usual ; and 

 it was pointed out in NATURE (m the notice cited above) 

 that one cause of these excessive deviations would be 

 found in the superposed lunar actions. Mr. Capello now 

 finds that a great majority of these quasi-solar disturb- 

 ances are rather to be considered due to the moon. 

 This conclusion induces me to believe that if Mr. Capello 

 had the necessary aid to perform the calculations for the 

 lunar diurnal variations for each month, and for different 

 positions of the moon, as well as for other investigations, 

 the Lisbon observations could not fail to add many im- 

 portant scientific results to those already published. 



John Allan Broun .' 

 Incidents in the Biography of Dust. By H. P. Malet. 



(London : Triibner and Co.) 

 The first impression one gets of this book is that of a 



