THE DARWINIA27 THEORY OF INSTINCT. 593 



This they continued to do, relieving guard, for three days, until they 

 had built supporting pillars of wax. Some other bumble-bees, when 

 shut up and so prevented from getting moss wherewith to cover their 

 nests, tore threads from a piece of cloth, and "carded them with their 

 feet into a fretted mass," which they used as moss. Lastly, Andrew 

 Knight observed that his bees availed themselves of a kind of cement 

 made of rosin and turpentine, with which he had covered some decor- 

 ticated trees — using this ready-made material instead of their own pro- 

 polis, the manufacture of which they discontinued ; and more recently 

 it has been observed that bees, " instead of searching for pollen, will 

 gladly avail themselves of a very different substance, namely, oat- 

 meal." Now, in all these cases it is evident that, if, from any change 

 of environment, such accidental conditions were to occur in a state of 

 nature, the bees would be ready at any time to meet them by intelli- 

 gent adjustment, which, if continued sufficiently long and aided by 

 selection, would pass into true instincts of building combs in new 

 directions, of supporting combs during their construction, of carding 

 threads of cloth, of substituting cement for propolis and oatmeal for 

 pollen. 



Turning to higher animals, Andrew Knight tells us of a bird 

 which, having built her nest upon a forcing-house, ceased to visit it 

 during the day when the heat of the house was sufficient to incubate 

 the eggs ; but always returned to sit upon the eggs at night when the 

 temperature of the house fell. Again, thread and worsted are now 

 habitually used by sundry species of birds in building their nests, 

 instead of wool and horse-hair, which in turn were no doubt originally 

 substitutes for vegetable fibers and grasses. This is especially notice- 

 able in the case of the tailor-bird, which finds thread the best material 

 wherewith to sew. The common house-sparrow furnishes another 

 instance of intelligent adaptation of nest-building to circumstances, 

 for in trees it builds a domed nest (presumably, therefore, the ances- 

 tral type), but in towns avails itself by preference of sheltered holes 

 in buildings, where it can afford to save time and trouble by construct- 

 ing a loosely-formed nest. Moreover, the chimney- and house-swal- 

 lows have similarly changed their instincts of nidification, and in 

 America this change has taken place wathin the last two or three hun- 

 dred years. Indeed, according to Captain Elliott Coues, all the species 

 of swallow on that continent (with one possible exception) have thus 

 modified the sites and structures of their nests in accordance with the 

 novel facilities afforded by the settlement of the country. 



Another instructive case of an intelligent change of instinct in con- 

 nection with nest-building is given from a letter by Mr. Haust, dated 

 New Zealand, 1862, which I find among Mr. Darwin's manuscripts. 

 Mr. Haust says that the Paradise duck, which naturally or usually 

 builds its nest along the rivers on the ground, has been observed by 

 him on the east of the island, when disturbed in their nests upon the 



VOL. XXTI. — 38 



