THE BIONOMICS OF AUSTRALIAN TABANIDAE. 53 



it was found that none were being devoured, when the water was drawn off and the 

 trough half filled with clean sand, which was kept moist. From time to time the sand 

 was turned over to watch the progress of development, but bevond this the larva 

 received no attention for intervals of several weeks. Some of the larvae pupated and 

 produced flies in this trough, but most of them were transferred to small pots of clean 

 moist sand, where they completed their development. In future it is intended to use 

 these large troughs or galvanized iron trays instead of the smaller vessels. 



Tabanus rufinotatus, Big. 



T. YHfinotatits, Bigot, Mem. Soc. Zool. France, v, 1892, p. 673. 



T. lineams, Taylor, Rept. Aust. Instit. Trop. Med. 1911, p. 65. 



T. elesteem, Summers, Ann. Mag. Nat. Hist. (8) x, 1892, p. 224. 



T. designatus, Ricardo, Res. Exped. Sci. Neerlandaise Nouv. Guinee, ix, pt. 3, 

 1913, p. 390. 



Distribution. This is a widely distributed species, having been recorded from 

 South Australia, New South Wales, Queensland and the Northern Territory, and also 

 from Dutch New Guinea. The South Australian specimens in the British Museum 

 collection are most probably from Port Darwin (Northern Territory) or its vicinity, 

 which prior to 1911 formed part of the state of South Austraha. 



Breeding-places. On 25th March 1919, an egg-mass was taken from the lower 

 surface of a Jnncea leaf growing three feet from the bank of a shallow water-hole 

 (near Locality 3) in twelve inches of water. At the time the pool contained a 

 maximum depth of three feet of water and was much frequented by cattle and 

 horses which grazed in the district. As the season advanced the water dried back 

 rapidly and finaUy disappeared before the end of May. During the period interven- 

 ing between 25th March and 28th May 1919 the vegetation near the water and the 

 muddy banks and bottom were searched for egg-masses and larvae respectively, 

 but without success. During the same period and up to the present date (September 

 1920) many other possible breeding-places have been repeatedly examined, but so 

 far only a few adult larvae have been secured. 



The Egg-mass. The egg-mass referred to above, from which the larvae described 

 in the following notes were derived, measured 5 • 5 mm. in length by 4 • 25 mm. in width 

 at the base and 3-5 mm. in height. The eggs, which numbered about 500, were 

 arranged very compactly in three tiers and were coated, either separately or collec- 

 tively, with a white secretion. After the larvae hatched, the mass retained its form 

 so perfectly that, viewed from any direction but from above, it appeared to be still 

 composed of viable eggs. 



Larval Development. The larvae hatched almost simultaneously at noon on 30th 

 March and moulted between that hour and 9 a.m. on the following day. Specimens 

 were not secured prior to ecd^'sis, so that the first instar cannot be described here. 



The Second Instar (figs. 9, 10). When 24 hours old the larvae of the second instar 

 measured from 2-5 mm. to 2-6 mm. in length by about 0-25 mm. in width at the 

 widest part. Up to this age they were of uniform size, but at the end of the second 

 day variations became apparent, some measuring 3 mm., while others had increased 

 to 3-5 mm. The cuticle is creamy white and bears minute longitudinal striae ; the 

 tufts at the base of the antennae are short and many of the component hairs are forked ; 

 the prothoracic segment is short, and the hairs which fringe its anterior margin and 

 the anterior margin of the next segment are difficult to discern. The dorsal tubercles 

 and pseudopods are armed with moderately large hairs, and each segment bears 

 several isolated long, slender, pale hairs. 



The young larvae remained in a small dish of water for three davs without food, 

 when they were placed in a large concrete trough containing sand at one end and 



