320 



NATURE 



\_Feb. 2 1, 1878 



dons. The two orders, Umbelliferte and AraliaccEe, both 

 by Mr. W. P. Hiern, and occupying the first thirty-two 

 pages, were printed in 1871 along with vol. ii., the rest 

 being all new. The district included in the present 

 volume is of course the same as that mentioned in the 

 previous ones, extending on each side of the equator for 

 somewhere about fifteen degrees. It is divided into six 

 regions, two on the west side of the continent, distin- 

 guished as Upper and Lower Guinea, two on the east 

 side ; Nile Land and Mozambique and the intervening 

 region divided into the north and south-central. The 

 district is therefore one of vast extent, and wi!l probably 

 yield many new forms when further exploration renders 

 our knowledge of the country and of its productions more 

 complete than it is at present. 



The Umbelliferas are represented in the flora by twenty- 

 one genera and about forty-four species. The number of 

 genera is small when contrasted with the thirty-four 

 found in Britain. Several familiar British forms are 

 found in this flora, of which it will only be necessary to 

 mention Sanicula ettropcea, Conium niaadatum, Anthris- 

 ais sylvesiris, Daucus carota, and Caucalis infcsta. Five 

 genera are, however, endemic in Africa. The genus 

 Petuedanum is interesting as including two species, P. 

 at-aliacen/n and P. fraxmifolhcm, both small glabrous 

 trees. The Araliacete are unimportant, and represented 

 by only three genera and fourteen species. 



By far the most important monograph in the 'present 

 volume is that on the Rubiaceae, by Mr. W. P. Hiern. It 

 occupies over 200 pages, or nearly as many as the mono- 

 graph of the Compositre, the joint production of Messrs 

 Oliver and Hiern. The Rubiacese form a very extensive 

 tropical and sub-tropical order, most richly represented in 

 America. Tropical Africa, however, possesses seventy- 

 eight genera, about thirty of which are endemic, and three 

 of these are now described for the first time by Mr. Hiern. 

 The genus Coffca is fully treated of by Mr. Hiern. 

 The C. arabica occurs in four of the districts of the 

 flora, but not in the two Central regions. The new 

 coffee, C. liberica, Hiern, which promises to be of so 

 much value, is here noticed. It is the source of the 

 Liberian coffee, and probably of the Cape Coast coffee. 

 The berries are said to be larger, the flavour finer, and at 

 the same time the plant is more robust and productive 

 than the ordinary C. arabica. The C. arabica has the 

 flowers pentamerous, while in C. liberica they are 7- or 

 6-merous ; or, according to the key to the species, 6 to 

 8-merous. The genus Sarcocephalus is an interesting 

 one, the fruits cohering to form a pseudocarp known as 

 the peach, or country fig, of Sierra Leone. The shape 

 and colour of the pseudocarp is that of a strawberry, but 

 in flavour it resembles an apple. Unfortunately if eaten 

 to excess it acts as an emetic. Two familiar species of 

 Galium are also included in the flora, namely^ G. aparine 

 and G. mollugo. 



The Valerianaceae are represented by the European 

 Valerianella dentata only. The Dipsacese by three genera, 

 Scabiosa yielding two familiar species. Both these small 

 orders are by Mr. Hiern. 



The Natural Family Compositas, the joint production 

 of Prof. Oliver and Mr. Hiern, occupies the chief place 

 in the volume, as would be expected of the largest order 

 of flowering plants. The number in the tropical region 



of Africa, 468 species, is small compared with the 1,300 

 species occurring in the Cape flora. In tropical Africa 

 there are 117 genera, seventeen being endemic, and all of 

 these latter either small or mono'.ypic. Many of the 

 forms are of great interest, and some of the genera very 

 extensive. It is curious to meet with Erigcroii alpinus 

 in Abyssinia, along with Senecio vuloaris, and others, 

 some of them familiar weeds. The genus Tarchonanthns 

 forms a small tree, and it was upon a species of this 

 genus that Dr. Welwitsch found the only species of 

 Visacin he met with in Angola. 



The Campanulaceae are by Mr. W. B. Hemsley, who 

 reduces the peculiar Abyssinian plants known as Tupa 

 to the genus Lobelia, and describes two species, L. rhyn- 

 chopetalum and L. giberroa. One species oi Lightfootia, 

 from Lower Guinea, is of interest. In it (Z. ivelwitschii) 

 the ovary is almost wholly superior, while in all others 

 the ovary is inferior, the plant thus looking very unlike 

 the other members of the Campanulaceee. Four genera 

 of Ericaceae occur, but only one species of Erica, viz., 

 E. arborea, thus contrasting with the Cape flora where 

 the species are most numerous. The EricaccE, Plumba- 

 gineas, and Primulaceas are by Prof. Oliver, but are small 

 and unimportant. The Myrsineas and Sapotacete are by 

 Mr. J. G. Baker, and the Ebenacea; by Mr. Hiern, who 

 has already published a monograph of this group. 



The greater part of the volume is by Mr. Hiern, who 

 contributes about 270 pages, while, along with Prof. Oliver, 

 he contributes 207 pages more. About twenty pages each 

 are contributed by Messrs. Baker and Hemsley, while 

 eight only are from the pen of Prof. Oliver alone. This 

 handsome volume of about 550 pages adds another to 

 the long series of " Floras " now so well known and so 

 highly appreciated that have from time to time issued 

 from Kew. W. R. McNab 



LETTERS TO THE EDITOR 



[The Editor does not hold himself responsible for opinions expressed 

 by his correspondetits. Neither can he tindcrtake to return, 

 or to correspond with the writers of, rejected manuscripts. 

 No notice is taken of anonymous couimunications. 



[ The Eddor urgently requests correspondents to keep their letters a r 

 short as possible. The pressure on his space is so great that it 

 is impossible otherzuise to ensure the appearance even of com- 

 munications containing interesting and novel facts. 'I 



Marine Fossils in the Gannister Beds of 

 Northumberland 



I TAKE the earliest opportunity to announce, through your 

 columns (if you will allow me to do so), that on the 9th of this 

 month (February), while conducting the usual weekly field-work 

 in connection with this college, I had the good foitune to fir,d 

 marine fossils in the lower coal measures, or gannister beds of 

 Northumberland. The locality is about half-way between the 

 Stocksfield Station, on the Newcastle and Carlisle Railway, and 

 the village of Whittonstali. As I have more than once insisted 

 on the fact that hitherto no marine forms had been found in this 

 series in this country, I wish to be the first to destroy the effect 

 which that negative evidence may have produced. The import- 

 ance which the study of the gannister fauna has assumed since 

 the publication of Prof. Hull's recent paper on the Classification 

 of the Carboniferous Rocks in the Quarterly Journal of the Geo- 

 logical Society, need not be dwelt on. 



Full details of the find will be published elsewhere. 



G. A. Lebour 



College of Physical Science, Newcastle-on-Tjne, February 12 



Liquids having a Specific Heat Higher than Water 

 In Nature, vol. xvii. p. 252, it is stated : " Hitherto 

 water has been regarded as possessing a greater specific heat 



