![Thumbnail: [Volume 86, 1959 page 269]](/journals/rsnz/images/rsnz_86/thumbnails/rsnz_86_03_0304_0269_ac_02.gif)
Introduction
The spiders included in the Symphytognathidae are all minute, ranging in body length from about 0.5 mm to 2.00 mm. Because of their small size and the cryptozoic life they lead in leafmould and moss, they have been rarely found in the past and are generally poorly represented in collections. However, over the last few decades sufficient numbers of these spiders have been collected and recognised to permit more conclusive studies to be made. Most of the specimens recorded in the present paper have been obtained over the last ten years from leafmould and moss by the use of Salmon's modification of the Berlese Funnel.
The family Symphytognathidae was established by Dr. V. V. Hickman in 1931 for the minute Tasmanian spider Symphytognatha globosa. The family was based on a number of unusual characters, of which the most important were the absence of lungbooks, the anterior spiracles leading into tracheae, the absence of female palps, and the fusion of the chelicerae along the midline. Petrunkevitch (1933) in his monumental study on the classification of spiders based on a study of their internal anatomy, pointed out that, in addition to the lack of lungbooks, the Symphytognathidae shared a number of internal characters with the Leptonetidae and Caponiidae, and he therefore placed these three families in a separate sub-order, the Apneumonomorphae. Fage (1937), after examining the respiratory system of a number of the spiders placed by Simon (1895) in the group Anapeae (Argiopidae), pointed out that all of the genera in this group which he was able to examine, with the exception of Tecmessa, were without lungbooks and should be in his opinion placed in the family Symphytognathidae, and he suppressed the family Anapidae established two years earlier by Kratochvil (1935) for these spiders. This conclusion has been followed by all subsequent authors.
Hickman, in the course of his series of papers on the spiders of Tasmania, established two further families of apneumone spiders, Micropholcommatidae (1944) and Textricellidae (1945), which he placed in the sub-order Apneumonomorphae with the Symphytognathidae, Telemidae and Caponiidae.
![Thumbnail: [Volume 86, 1959 page 270]](/journals/rsnz/images/rsnz_86/thumbnails/rsnz_86_03_0305_0270_ac_02.gif)
After studying the large series of spiders now available from New Zealand, Australia, New Guinea and the Pacific Islands, and the collection of these spiders from North and South America in the American Museum of Natural History, New York, I have concluded that the families Symphytognathidae, Micropholcommatidae and Textrcellidae, with the genera Mysmena and Lucharachne could well be placed in a single family, for which the oldest available name is the Symphytognathidae.
In spite of our increasing knowledge of the structure and habits of these spiders the relationship of this family to other spiders still remains in doubt. The genera included in the wider interpretation of the family adopted in this paper could be grouped equally well on morphological grounds to demonstrate close affinity with either the Argiopidae or the Therdiidae. Opinion in the past has been divided in the main between these two alternatives. Simon (1895) placed his group Anapeae at the end of the Argiopidae, indicating affinity with the orbweb spiders, while Berland (1924) considered that the group should be placed in the Theridiidae. Fage (1937), while following Simon's placing of the group, pointed out that only future study would resolve the question of whether the family had evolved directly from either the Argiopidae or Theridiidae or originated from stock common to both of these families. I am inclined to the view that the family has been derived from the Argiopidae or at least has evolved from a stock common with the Argiopidae. The fact that a number of genera construct typical orbwebs—Risdonius (Hickman, 1938), Chasmocephalon (Hickman, 1946), Patu (Marples, 1956)—in my opinion strongly supports this view. It is most difficult to conceive the separate development of an orbweb in a form identical with that of typical argiopid spiders, whereas degeneration from an orbweb could easily lead to webs having the appearance of sheet or tangle webs. Textricella constructs small sheet webs which look like those of the Micryphantidae while the webs of Micropholcomma are tangle webs such as are found in the Theridiidae. Archer (1946) reports that North American species of Mysmena constructs a sheet web, but the statement made by Marples (1955) after carefully studying Tamasesia acuminata, a species which is transferred to Mysmena in the present paper, seems significant. He states: “The web is extremely delicate. It consists of a set of threads radiating in all directions from a centre where the spider sits. The space between the radials is filled with the threads of sticky silk, so fine that the droplets can only be seen under the microscope, and the whole occupies a volume roughly 1.5–2 cm across. When spinning the spider keeps going quickly out along different radials. Apparently it attaches a thread to a radial and carries the other end to the centre and out along another radial to attach it there. The web is built from the periphery inwards. Though the threads are not regularly arranged, the general impression is of an orb web in three dimensions.” If we were attempting to bridge the gap between the typical orbwebs found in Chasmocephalon, Risdonius and Patu and the type of web found in Micropholcomma, the web of Mysmena acuminata provides a perfect example. With only slight modification this type of web would lose all resemblance to an orbweb, and would then be in the form found in Micropholcomma. With reference to the development of a sheet web from an orbweb, I might quote Marples' (1951) reference to the web of Patu samoensis, where he states: “They consisted of a very fine horizontal sheet with very regular meshes. The sheet was an irregular polygon some six centimetres long, and an oblique thread extended upwards from the sheet between one and two centimetres from one end. The threads of the sheet radiated from a point of attachment of the oblique thread, but it was a sheet and not an orb web.” Later observations, however (Marples, 1955) demonstrated that this web is initially an orbweb of very fine mesh, and that subsequent tangling of the threads or the addition of further threads gives it the appearance of a sheet web. During a brief stay in Fiji in 1956 I was able to examine the web of the very closely related species Patu vitiensis. The webs, which are from 3 to 5 centimeties in width, were on tree trunks, where they were placed horizontally. The threads were extremely fine and
![Thumbnail: [Volume 86, 1959 page 271]](/journals/rsnz/images/rsnz_86/thumbnails/rsnz_86_03_0306_0271_ac_02.gif)
closely spaced, but after treatment with talc the web was found to be a perfect orb, with the sticky spiral lines closely spaced, and it does not appear that in this species the regular structure is subsequently modified as in samoensis.
Symphytognatha is reported by Hickman (1931) to construct an irregular web like that of Theridion, in which the spider rests in an inverted position. In view of the very close morphological relationship of this genus to Patu it would be of considerable interest if the method of web construction were to be closely studied to see if there is any evidence for this web being a degenerate orbweb.