March 4, 1886] 



NA TURE 



415 



the memoir published by me in 1S73. Lister's observations led 

 him to quite different conclusions, which he has since aban- 

 doned. I am sure that those who are at present busy in this 

 country with the study of Bacteria, and who undertake to write 

 hand-books of the subject, can have do desire to do otherwise 

 than give a just statement of the history of knowledge of the 

 organisms of which they treat. Hence it is with no unfriendly 

 feeling that I offer to Mr. Crookshank and other writers 

 similarly engaged the statement contained in this letter. 



February 20 E. Ray Lankester 



Notes on the Volcanic Phenomena of Central 

 Madagascar 



M.ADAGASCAR is as yet almost a tirra incognita to the geolo- 

 gist ; nothing, so far as I am aware, but notices of the most 

 vague and fragmentary kind tver having appeared in regard to 

 its geological features. Nor indeed may we e.\pect to have 

 other than the most general descriptions until the island is sur- 

 veyed by thoroughly competent men. In the absence of some- 

 thing more complete, I hope that the following notes on the 

 volcanic phenomena of Central Madagascar may not be un- 

 acceptable to your readers, and may prove a contribution, how- 

 ever slight, to our knowledge of the geology of this great island. 

 And first let me mention the volcanic cones, of which there are 

 many scores, probably hundreds, in the part of the island of 

 which we are speaking. These volcanic cones are situated in 

 two localities especially : in Mandridrano, on the western side 

 of Lake Itasy, and in the neighbourhood of Betafo in Vakin' 

 Ankaratra ; the former being from 50 to 60 miles west, and the 

 latter from 70 to So miles south-west, of Antananarivo, the 

 capital. Both localities are about 130 miles from the sea on the 

 eastern side of the island, and 150 on the western side. It is 

 hardly necessary to say that all these volcanoes are extinct, and 

 that there are none in activity at the present time in any part of 

 Madagascar.^ On the west side of Itasy the volcanic cones 

 e-tist in great numbers, and these, therefore, shall be first 

 described. 



The extinct volcanoes of this district of Mandridrano extend 

 for a distance of about 20 miles north and south, and perhaps 

 3 or 4 east and west. They are, for the most part, scoria cones. 

 The cones are thickly studded over the district, in some parts 

 clustering together more thickly th.in in others. There is no 

 single large volcano to which the others are subsidiary, or upon 

 which they are parasitic. Occasionally there is a series of cones 

 which have evidently been heaped up by the simultaneous ejec- 

 tion of scorise from different vents situated on the same line of 

 fissure, but so that the cones have run one into the other, leaving 

 a ridge, generally curvilinear, at the summit. None of these 

 extinct volcanoes reach the height of loco feet. Kasige, which 

 is probably the highest, I found by aneroid to be 863 feet above 

 the plain (5893 feet above the sea). Andranonatoa is perhaps 

 next in height to Kasige. Kasige is a remarkably perfect and 

 fresh-looking volcano, whose sides slope at an angle of about 

 40°. The scoriEe on the sides have become sufficiently disin- 

 tegrated to form a soil on which are found a by no means scanty 

 flora ; for among other plants growing here I found an aloe 

 [A. macioclada), and clematis (C trifiJd), two or three Com- 

 posite herbs [Semcio cochharifoliu^^ Hclichrysuvi lycopodioidis^ 

 Laggira alata, &c.), some grasses (Impcrata arundinacca, &c.), a 

 species of Indigofera, and an orchid. On its top is an un- 

 breached crater, which measures, frjm the highest point of its 

 rim, 243 feet in depth. It may be mentioned in passing, that 

 on the very summit, in a hollow "cinder," there were found 

 a small piece of money, perhaps of the value of a halfpenny, 

 and a small bead, as also a portion of a bai ana leaf, with a few 

 pieces of a manioc, and two or three earth-nuts placed upon 



' Scrope, in his *' Volcanoes," second edition, p. 42S, says of Madagascar, 

 " There is scnie reason to believe in the existence of active volcanic vents 

 in this great island ; " and Dr. Daubeny, in the second edition of his '' Vol- 

 canoes," p. 433, in refening to the islands on the eastern coast of Africa, 

 says ; ■' The principal of these are the great Island of Madagascar, the 

 Isle of Bourbon, and the Mauritius, the first of which has been too little 

 explored to allow of my announcing with certainty anything respecting its 

 physical structure ; " and in a note he adds ; '* Madagascar is stated by 

 Daubuisson to contain volcanoes, on the authority of Ebel i^Bau dcr Erde, 

 tome ii. p. 289), who reports that in this island there is a volcano ejecting a 

 stream of water to a sufficient height to be visible 20 leagues out at sea." 

 What remarkable eyesight those from whom Daubuisson heard the story 

 must have had to see an invisible phenomenon so far away ! Or. Daubeny 

 continues: "Sir Roderick Murchison, December 1827, exhibited at the 

 Geological .Society some specimens of a vjlcanic nature, said to h.ive come 

 from this island, but the locality was not mentioned." 



it these had been deposited there by some of the heathen 

 inhabitants of the place as a votive offering either to their 

 ancestors or to the Vazimba (the aborigines of Central Mada- 

 gascar). Continuous with Kasige, and adjoining its south side, 

 though not so high, there is another volcano, Ambohimalala, and 

 dozens of others are to be seen near by. 



One thing with regard to these volcanic piles soon strikes the 

 observer, which is, that they are frequently lop-sided, one side 

 of the crater being higher than the other. The higher side 

 varies from north to north-west and west. This is undoubtedly 

 accounted for by the direction of the wind during the eruption, 

 causing the ejected fragments to accumulate on the leeward side 

 ot the vent. Now we know that the south-east trades blow 

 during the greater part of the year in Madagascar, hence the 

 unequal development of the sides of the cones. The same 

 thing may be also observed in the volcanic piles in the neigh- 

 bourhood of Betafo. This phenoinenon, as is well known, 

 occurs also in other parts of the world. 



A very large number of the cones have breached craters, 

 whence lava has flowed in numerous streams and floods, covering 

 the plains around. These streams and floods consist in every 

 instance, I believe, of black basaltic lava ; a sheet of this lava, 

 the mingled streams of which have flowed from Ambohimalala 

 and soine other vents, has covered the plain at the foot of Kasige 

 to such an extent as almost to surround the mountain. Similar 

 sheets are to be seen in other parts of the district, but they are 

 so much alike that a description of one will suffice for all. 

 Amboditaimamo (or Ambohidratrimo ?) is a small volcano to the 

 north of Lake Itasy, and at the northern confines of the volcanic 

 district. It possesses a breached crater turned towards the east ; 

 from this has issued a stream of lava which, following the 

 direction of the lowest level of the ground, has swept through 

 a small valley round the northern end of the mountain, and 

 spread out at its west foot. This sheet of lava, which is horribly 

 rough on the surface, occupies but a small area of some two or 

 three square miles. It has been arrested in its flow in front by 

 the side of the low hills. It is cut through in one part by a 

 stream which, in some places, has worn a channel to the depth 

 of 80 or 90 feet. Its surface, which is slightly cellular, is 

 covered by some hundreds of mammiform hillocks, which must 

 have been formed during the cooling of the liquid mass. The 

 hillocks are mostly from 20 to 30 feet high, and apparently are 

 heaped-up masses of lava, and not hollow blisters. The lava 

 itself is black, heavy, and compact, being porphyritic with 

 somewhat large crystals of augite. As yet it is scarcely decom- 

 posed sufficiently to form much of a soil, though grass grows on 

 it abundantly, and a few other plants are to be seen. 



A little to the south of Ainboditaimamo there is another vol- 

 cano, known by the name of Andrarivahy. It is situated on 

 the summit of a ridge of hills — astride of it, so to speak — and 

 from its crater there has been an outflow of what must have 

 been very viscid lava, for, though the sides of the volcano and 

 the ridge of hills form an angle of from 30" to 40°, the ejected 

 matter has set or "guttered" on the slope, only a small portion 

 of it having reached the valley below. This ridge of hills, 

 through which the volcanic orifice has been drilled, is composed 

 entirely of gneiss ; and indeed it inay be here stated that the 

 whole of these volcanoes, as is the case also with those about 

 Betafo, rest upon a platform of gneiss, chiefly garnetiferous. 



Throughout the district numerous fraginents of basic lava, 

 trachyte, trachytic tuft', and basaltic conglomerate lie scattered 

 about in abundance. The trachyte is of various shades of yellow 

 and gray, and frequently porphyritic wilh large crystals of sani- 

 dine. Pumice, obsidian, and pitchstone do not seem anywhere 

 to be found. 



In addition to the nuinerous scoria-cones, there may be seen 

 here and there in the district some half-dozen or more bell- 

 shaped hummocks of trachyte. They are for the most part 

 coinposed of a light-coloured compact rock. This rock, having 

 originally had a highly viscid or pasty consistency, has evidently 

 accumulated, and set immediately over the orifice through which 

 it was extruded; such hummocks are Ingolofutsy, Beteheza, 

 Angavo, .\mbasy, Isahadimy, Ambohibe, Antsahondra, &c. 

 Ingolofolsy, .'■ituated to the north-west of Itasy, is perhaps the 

 most striking in appearance of these trachytic hummocks. It 

 bears some resemblance to a bell or Turkish fez, except that its 

 sides are furrowed with watei-channels and its truncated summit 

 is notched in a remarkable manner. Its height above the plain 

 is 665 feet (525S feet above the sea); the inclination of its sides 

 averages probably 50^ Adjoining Ingo'ofo sy on the south- 



