VIRGIN FOREST. 27 



Another interesting physical feature of Madagascar, which 

 has only been made clear very recently, is the existence of an 

 almost continuous belt of virgin forest all round the island, 

 and generally following the coast-line. This forest divides 

 into two belts on the eastern side of the country, leaving a 

 long narrow valley about 250 miles long between the two 

 lines. The uppermost of these clothes the slopes which form 

 the edge of the upper plateau of the island. North of this 

 valley the two lines unite, and here is the widest portion of 

 the forest, it being about forty miles across. The average 

 breadth is from fifteen to twenty miles. On the north-west side 

 the two lines overlap each other nearly 100 miles, leaving an 

 opening about seventy miles wide. The total length of this 

 forest must be about 2300 miles, and much of this is yet un- 

 explored, so that there is doubtless still much of interest in 

 botanical science awaiting research. Besides the forest belt, a 

 good deal of the country to the west and south is well wooded. 



A third fact of interest in the physical geography of Mada- 

 gascar is the extensive evidence of recent volcanic action 

 throughout a great part of the country. It has been known 

 for several years that there were signs of this on the north- 

 west coast, and that in the island of Nosibe and the adja- 

 cent mainland there are numerous extinct craters and much 

 igneous rock. A few years ago the Eev. T. Campbell, of the 

 Church Missionary Society, pointed out evident traces of 

 volcanic agency in the district near the Ankaratra hills. He 

 says : " It seemed as if the whole place were once a great 

 smeltery, from the enormous number of clinkers lying about. 

 There were altogether five mountains, all near to each other, 

 which have been active volcanoes at some remote period; 

 each has one of its side^ melted down, and the inside hollowed 

 out. The flow of lava looks as if it had been some immense 

 reservoir bursting its banks, and the water dashing and foam- 

 ing through, bearing every tnmg away with it, or covering the 

 plain beneath." 



In a journey I took to the lake It^sy in 1866, I was 

 struck with the number of truncated cones in the hills sur- 

 rounding the lake. But extensive journeys made more 

 recently in various directions have revealed the existence of 



