512 TRANSACTIONS OF SECTION I 
wall of the luminated portion of the cleft gives origin to epithelial body I1I., 
whilst the whole of the caudal wall of the same, as well as the solid ventral 
portion of the cleft (which rapidly grows downwards beyond the connection with 
the ectoderm of the cervical sinus) turnishes the epithelial basis of thymus I1T. 
The fourth gill-cleft undergoes a corresponding development to cleft 111., only 
here the epithelial body (1V.) 1s derived from the dorsal part of the cleft, includ- 
ing the upper parts of both its cranial and caudal walls, whilst thymus LY. takes 
origin exclusively from its more ventral portion. 
The possession of a fully developed thymus derived from the fourth cleft is a 
primitive feature characteristic (so far as at present known) of the Marsupials 
alone amongst the Mammals. Rudiments of a thymus LV. have, however, been 
observed in a number of EKutheria, whilst a thymus IV. is stated to occur in 
some reptiles, e.g., snakes. In respect of the presence of a distinct thymus LV. 
as well as in the mode of origin of thymus Ill. from the entire caudal as well 
as the ventral wall of cleft LiI., Zvichosurus exhibits more primitive relations 
than any Mammal hitherto investigated. 
The thyroid would appear to be of composite origin. Its main mass takes 
origin from a median ventral outgrowth of the pharynx arising in the normal 
position between clefts II. and ILI., but it is highly probable that the post- 
branchial bodies also take part in the formation of the lateral lobes. The post- 
branchial body on each side after separating from the fourth cleft comes to 
lie on the mesial side of the thyroid lobe, and the two become enclosed in the 
same capsule. From the walls of the post-branchial body there arise sprout-like 
processes which penetrate amongst, and become indistinguishable from, the 
cellular cords of the lateral lobe. It is probable that these processes also give 
rise to thyroid vesicles. 
9. Fat-tailed Sheep. By Professor J. C. Ewart, F.R.S. 
P y ‘ 
About fat-tailed sheep one naturally asks, ‘From what wild ancestors are they 
descended, and for what purpose is fat stored up in the tail?’ There is no 
evidence that long-tailed sheep, with or without a store of fat, are descended from 
long-tailed, wild ancestors. ‘Three types of wild sheep exist at the present day. 
There are sheep of the Urial (Ovis vigenci), of the Mouflon (Ovis orientalis), and 
of the Argali (Ovis ammon) type. 
All true wild sheep are characterised by a short tail. In the Peat sheep (Ovis 
palustris), which reached Europe in Neolithic times, we have a domesticated 
variety of the Urial—a short-tailed race (found on Soay, near St. Kilda, and in 
Shetland), with goat-like horns in the female, which is perhaps the best living re- 
presentative of the Peat sheep. In Studer’s sheep (Ovis studeri), of Copper Age 
deposits, we have a domesticated variety of the Moutlon—a short-tailed race, the 
males with large curved horns, but the females hornless, which occurs in Soay 
and Shetland, and seems to be the best living representative of the Copper Age 
sheep. In the short-tailed fighting ram of Nepal, with large spiral horns, we 
seem to have a somewhat modified representative of the Argali, or of its ally, 
Ovis poli. Some fat-rumped sheep with a short tail (e.g., fat-rumped Persian 
sheep) may be descended from domesticated varieties of the Urial or Mouflon, 
but sheep with a long fat tail (¢.g., the fat-tailed sheep of Afghanistan) are 
probably descended from domesticated breeds allied to the fighting ram of Nepal. 
It is highly probable that as the large inland seas, common in Central Asia 
in prehistoric times, dried up, domesticated sheep, to have a chance of surviv- 
ing, found it necessary to store up fat by way of providing nourishment during 
the long dry season. 
In some cases fat was deposited to form fat-rumped races, in others to form 
fat-tailed races. The individuals which, by increasing the number and length of 
the tail vertebre, provided most accommodation for fat would, in the struggle 
for existence, have the best chance of surviving, as the aridity in Central Asia 
increased. 
It is extremely probable that the long-tailed European breeds, instead of 
inheriting, as used to be assumed, their long tails from an extinct long-tailed 
wild ancestor, are indebted for their long and apparently useless caudal append- 
ages to fat-tailed ancestors. Evidence in support of this view is afforded by the 
a 
