442 PHYSICAL GEOGRAPHY. 



it, it waits exactly the length of time it would have taken for it to perfect 

 its forward progression, when it returns on its path again. But this takes 

 place only with such species as have a regular backward and forward mo- 

 tion ; most of them are extremely erratic in their ways. As an illustration 

 of the regularity of movement of the diatomacese, let us consider one of 

 those species just mentioned, namely, Bacillaria paradoxa. This crea- 

 ture has a motion of its own, which is so peculiar, and at the same time 

 so incomprehensible in its mode of accomplishment, that it well deserves 

 a more particular description. This species is somewhat of the form of 

 a straight ruler, when we consider the single individual, or, at least, it 

 looks very much of that form, as commonly seen. But a careful exami- 

 nation shows it to be made up of two long and narrow boat-shaped valves, 

 united together so that the keels project outwards opposite to each other, 

 and enclose within their united bodies the general cavity of the frustule. 

 When in a living state, the compound individual, or colony, whichever 

 we may choose to consider it, consists of a number, more or less great, 

 of these double boat-shaped frustules, united one to the other at their 

 keeled sides ; but the mode of union is entirely unknown, and, from the 

 extreme freedom of motion which each frustule of the colony enjoys, it 

 is hard to imagine what its character is. Thus united, they form a fila- 

 ment which is generally found floating freely in the water of brackish 

 ditches within reach of the influence of the sea. But I have seen it in 

 perfectly fresh water, far up the Hudson river in New York state. 



The movement has been well described by an English observer, Mr. 

 Thwaites, and I cannot do better than quote his words. He says, — 



When the filaments have been detached from the plants to which they adhere, a 

 remarkable motion is seen to commence in them. The first indication of this consists 

 in a slight movement of a terminal frustule, which begins to slide lengthwise over its 

 contiguous frustule ; the second acts simultaneously in a similar manner with regard to 

 the third, and so on, throughout the whole filament ; the same action having been going 

 on at the same time at both ends of the filament, but in opposite directions. The cen- 

 tral frustule thus appears to remain stationary, or nearly so, while each of the others 

 has moved with a rapidity increasing with its distance from the centre, its own rate of 

 movement having been increased by the addition of that of the independent move- 

 ment of each frustule between it and the central one. This lateral elongation of the 

 filament continues until the point of contact between the contiguous frustules is 

 reduced to a very small portion of their length, when the filament is again contracted 



