FOOD OF INSECTS. 405 



end of a pin, is furnished, according to Reaumur*, 

 with a thousand of them. From each of these tubes, 

 consisting- of two pieces, thelastof which terminates in 

 a point infinitely fine, proceeds a thread of inconceiv- 

 able tenuity, which, immediately after issuing- from it, 

 unites w ith all the other threads into one. Hence from 

 each spinner proceeds a compound thread; and these 

 four threads, at the distance of about one-tenth of an 

 inch from the apex of the spinners, again unite, and form 

 the thread we are accustomed to see, which the spider 

 uses in forming its web. The threads, however, are 

 not all of the same thickness, for Leeuwenhoek ob- 

 served that some of the tubes were larger than others, 

 and furnished a larger thread. Thus a spider's thread, 

 even spun by the smallest species, and when so fine that 

 it is almost imperceptible to our senses, is not, as we 

 suppose, a single line, but a rope composed of at least 

 four thousand strands. How astonishing! But to feel 

 all the wonder of this fact we must follow Leeuwen- 

 hoek in one of his calculations on the subject. This 

 renowned microscopic observer found by an accurate 

 estimation that the threads of the minutest spiders, 

 some of which are not larger than a grain of sand, are 

 so fine that four millions of them would not exceed in 

 thickness one of the hairs of his beard. Now we know 

 that each of these threads is composed of above 4000 



' Reaum. Mem. dc VJcad. de Paris, An. 1713. 211.— De Geer, vii. 187. 

 See also Hoole's Leeuwenhoek, i. 41. t. 2. /. 20-22. Leeuwenhoek ex- 

 amined a spinner that was not so big as a common grain of sand, and the 

 number of tubes issuing from it was more than a hundred. He affirms 

 that, besides the larger spinners, in the space between them there are four 

 smaller ones, each furnished with organs for spinning threads but smallep 

 and fewer in number. See Plate XXIII. Fig. 16, 17. 



