Febkuarv 28, 1895 J 



NA TUKE 



419 



will presently be explained, that in both cases it has been 

 brought to its present state of perfection under stress of 

 the same adverse conditions of life. As is well known, 

 Mr. Moggridge long ago suggested that the instinct to 

 construct the door may have arisen from the habit of 

 closing the aperture of the tube in the winter and open- 

 ing it again in the spring. This idea, in substance, has 

 been adopted and further developed by Dr. McCook, 

 who states, upon the authority of Mrs. Treat, that a North 

 American species of wolf-spider [Lvcosa //^r/w,?) has 

 acquired the instinct of sealing up the aperture of her 

 nest during the breeding season of the Mason-wasps ; 

 for at this period these insects scour the country for 

 spiders, in order that they may lay up a store of food for 

 their young. When the wasps have disappeared with 

 the close of their hunting and breeding season, the 

 spiders venture again to remove the covering of their 

 nests ; but Mrs. Treat has made the further important 

 observation, that some examples leave the covering 

 attached at one point. Thus a genuine, though roughly- 

 formed, trap-door nest is produced. In view of this cir- 

 cumstance, there cannot be much doubt that the per- 

 manent and highly-finished trap-door nest of the Russian 

 Lycosa opifex has been similarly brought about, as M. 

 Wagner, the discoverer of the species, has suggested, 

 under the stress of the dire persecution from wasps to 

 which spiders in general are subjected. 



Being thus able to trace with some degree of certainty 

 the steps by which the trap-door nest has been evolved 

 in one group of spiders, namely, the Lycosidtc, we are 

 justified in concluding, at all events until evidence to the 

 contrary is forthcoinLng, that it has been evolved in the 

 same way in the case of the Aviculanithc — the trap-door 

 spiders par cxcelloite. 



The primary influence, then, that has been at work in 

 guiding the evolution of the architecture of the tunnel- 

 making species, has apparently been that great necessity 

 for the preservation of life, the avoidance of enemies, 

 liut if we turn to the other line, along which the web- 

 building instinct has been developed, we find that the 

 primary guiding influence has been that second great 

 vital necessity, the acquisition of food. 



As has been already stated, the origin of the webs 

 which function as snares seems to be referable to a 

 simple silken tent or tube, similar to that from which all 

 the niore or less complicated forms of tubular nests 

 appear to have been developed. Perhaps the most rudi- 

 mentary form of snare arose, as Dr. McCook has 

 suggested, from the chance spinning of a few stray 

 threads about the mouth of the tubular retreat ; or, 

 perhaps, an irregular network of threads spun around the 

 aperture to interfere with the entry of such enemies as 

 wasps, was the first step in the evolution of net-spinning ; 

 nr even lines anchoring the tube securely in its site 

 might have first served the purpose of catching prey. 

 liut, however this may be, it is clear, as Dr. Romanes' 

 has poined out, that "there is much potential service 

 to which the power [of net-spinning] may be put with 

 reference to the voracious habits of the animal. ' Taking 

 this into consideration with the variation in structure 

 presented by different species of spiders, it is not sur- 

 prising that there are many modifications of the net. 

 Sometmies it is a thick, closely-woven horizontal sheet, 

 which is continuous at one extremity with a tubular 

 retreat, as in the case of one of our commonest house- 

 spiders, Tc'i^enaria ; or, as in the equally common Aiiuiu- 

 robius, the net is less regular in shape and less thickly 

 woven, but is still continuous, with a silk-lined hole, in 

 which the spider lurks ; or again, the web, as in P/io/ius 

 or 'J /leriitiinii, may be composed of an irregular mesh- 

 work of interlacing threads, without any such tubular 

 retreat as that constructed by Ta^icnana or AiiuxKrobius ; 

 or, lastly, it may be composed of radiating and concentric 



' "Animal Intelligence," p. aoS. 

 NO. 1322, VOL. 51] 



lines, like that of our garden spider, Epeira : and it 

 seems to be generally admitted that this orbicular web of 

 Epeira manifests the greatest perfection of instinct, and 

 is therefore to be regarded as the highest form of this 

 kind of spinning-work. Consequently, the question con- 

 cerning the possible steps by which such a structure has 

 been evolved cannot fail to be of interest. 



In the first place, if all snares are traceable back to a 

 common tubular origin, it may be taken for granted that 

 those that are still associated with a tubular retreat are,, 

 cateris paribus, of a more primitive type than those in 

 which the tube has been abandoned. Furthermore, it may 

 be confidently assumed that the habit of weaving the lines 

 of the snare radially and concentrically in a definite and 

 elaborate pattern, was preceded by the habit of arranging 

 them irregularly and without order. Looked at from this 

 point of view, the web of a Te^enaria or Amaurobiiis is a 

 much less specialised structure than that of an Epeira. 

 It may consequently be concluded that the complete 

 orbicular snare of the latter animal, and of orb-weavers 

 in general, has been derived from one which, like that of 

 the tunnel-weavers, ivas composed of irregularly crossing 

 threads, and was continuous at one extremity with a 

 tubular domicile. Having arrived at this conclusion, we 

 naturally appeal to nature for corroboration, and search 

 for connecting links. Nor need we look far. For, taking 

 first the tunnel-weavers, we find that a species of Dictyna, 

 a spider nearly allied to our common Amaurobiiis, con- 

 structs a snare of which the threads are arranged radially 

 and concentrically, but so roughly that the resemblance 

 to the finished structure with which we are familiar in 

 our garden-spiders is only remote. Nevertheless, one 

 cannot avoid the conclusion that it represents an initial 

 stage in the developinent of the perfect orb. 



Turning, in the next place, to the orb-weavers, we 

 naturally look out for snares constructed upon a more 

 primitive plan than that which is typical of our English 

 species of Epeira. But if there be any such in existence,^ 

 we should reasonably expect, in accordance with our 

 hypothesis, to find these simpler kinds associated with a 

 tubular retreat. And our expectation would be justified 

 by facts. For the large and handsome tropical genus 

 Nephileni^ys spins a web which is structurally interme- 

 diate in character between \\rA\.Ci{ Epeira diadctnala{s>\ix 

 garden-spider) and that of the tunnel-weaver, Die/yna.^ 

 This web resembles that of Tegeiiaria and Dictyna, in 

 consisting of a long silken tube, with an expanded 

 funnel-shaped mouth opening directly upon an extended 

 network of threads. But the latter, instead of being 

 fashioned like that of the majority of tunnel-weavers, 

 consists of a scanty mesh-work of lines arranged radially 

 and concentrically with respect to the mouth of the 

 funnel. In this particular it is similar to the net of our 

 garden-spider, Epeira; but its area, instead of forming 

 a complete circle, extends over only about one-third of 

 this figure. The importance, however-, of this distinction 

 breaks down when the webs of other species of orb- 

 weavers are taken into consideraton. For it is found 

 that those of the Malaysian Epeira bcccarii, as figured 

 by Mr. Workman, and of the North American Epeira 

 lahyriiitliea of Hentz, are completely circular, and yet the 

 radial threads at the centre of the web spring from the 

 mouth of a long silk tube, in which the spider lurks. 



To all intents and purposes, therefore, there are not 

 many links missing in the chain which starts with the 

 web of a tunnel-weaver, like our house-spider Tegenaria, 

 and terminates with that of our garden-spider Epeira. 

 Furthermore, from the web of Tegenaria gradations 

 may be traced backwards to the simple tubular retreat 



1 I have to th.tnk my friend Mr. H. \. Spencer for sketches of the web ot 

 •a species of this spider, and .-ilso for a living example of theaitinial which he 

 kindly brought to me from Durban, while acting as medical oflicer on 

 board the s.s. .l/r.i/Vaw. I was forttinalc enough to keep this spider alive 

 for several months, and was thus enabled by person.al observation to satisfv 

 myself of the accuracy of Mr. Spencer's representation of the web. 



