260 D. Prain — The non-indigenous species of the Andaman Flora, [No. 3, 



spontaneously and appearing likely as time goes on to increase perceptibly 

 the numbers of the non-indigenous flora. The greater number of these 

 must have certainly been introduced in the 1789-92 period, and many of 

 them are such as at first sight suggest for themselves the possibility of 

 survival. 



Perhaps, however, it ought not to surprise us greatly that species 

 which readily appear spontaneously elsewhere and which are appearing 

 spontaneously in the Andamans now, should, if they were px*eviously in- 

 troduced, have perished between 1792 and 1858. Most of them are plants 

 that, when they do escape from cultivation and appear spontaneously, 

 affect such situations as waste places, rubbish heaps, road-sides, hedge- 

 rows and margins of clearings, — situations that have at least this in 

 common, that they afford their denizens abundance of air and light. 

 Many of them too are herbaceous, or at most fruticose, and the native 

 jungle as it reinvades the abandoned clearings overshadows them and 

 either chokes them completely, or by merely preventing them from 

 flowering*, makes their fate only a matter of time. Even trees that seem 

 quite naturalised in clearings must soon succumb to the weight of creepers 

 that rapidly overload them in a forest. 



If, however, the survival of even a small proportion of the cultivated 

 species abandoned in 1792 will suffice to explain the higher rate of na- 

 turalisation during Period I, deducible from the figures in Table III 

 (Oarica Papaija and Gocos nucifera are excellent examples of such sur- 

 vival), there is no similar explanation possible for the higher rate of weed- 

 introduction during the same period. A considei*able number must have 

 been already introduced by 1792, and, though many doubtless yielded to 

 the influences adverse for naturalised cultivated species, weeds are often 

 proverbially tenacious of life and a good few, as the notes against them 

 show, in place of avoiding the jungle are actually penetrating into it. 

 Taken altogether we find that the rate of introduction during the first 

 period was 2| times as high as it has been during the second, and the 

 most probable explanation of this higher early rate of weed introduction 

 appears to be that in the dirty grain of an Indian bazaar seeds of most 

 of the commoner Indian weeds are certain to be present. This being 

 the case so many weeds become introduced with the very earliest sowings 

 of any grain that the subsequent rate of introduction of species can be but 

 small. And it is highly probable that for the same reason the rate of weed- 

 introduction becomes year by year diminished. Unfortunately it has not 

 occurred to any one to make observations on these weeds during the 

 interval 1858-66 or 1866-90. And without repeated observations after 

 short intervals of time, especially towards the commencement of a 

 settlement, it is impossible to test the adequacy of this explanation. 



