DISTRIBUTION BRITISH ISLES 607 



"A remarkable feature of many of them is their subdivision into large, 

 irregular, sack-like or pillow-shaped blocks, which may have their central 

 portions more largely vesicular than the rest. These ellipsoids were formed 

 during the flow of the still moving lava along the floor of the lagoon or sea. 

 The interstices between them have often been filled in with fine tuff, which 

 was stratified horizontally from side to side in the fissures before the lava 

 was covered by the next outflow of molten material. In some cases portions 

 of the still fluid lava in the heart of the mass were forced into the interstices 

 and now appear as veins of finely cellular basalt. The lava sheets range 

 from 8 or 10 to 40 feet or more in thickness" (pages 54, 55). 



"This well-marked 'pillow structure' appears to indicate that, while still 

 moving as molten masses, they separated into irregular ovoid, sack-like or 

 pillow-shaped blocks of all sizes, from less than a foot to 5 or 6 feet in 

 length. These blocks are frequently most cellular in the center, the vesicles 

 being there largest in size and most crowded together. In other cases the 

 vesicles are grouped around the margin and sometimes more particularly 

 along one side of each ellipsoid. In the interstices between the blocks fine 

 tuff and ashy sandstone may sometimes be seen, showing that the rude bowl- 

 der-like masses were more or less separated from each other, so as to allow 

 fine sediment to be dropped into the interspaces. This sediment is stratified 

 horizontally or in the same direction as the general bedding-plane of the lava 

 in which it lies. 



"As they rolled along over the lagoons and pools of the time the basalts 

 now exposed on the shore west of Pettycur caught up and involved large 

 quantities of the muddy and calcareous sediments which lay in their way" 

 (page 62). 



In the southern borders of the Highlands, says Geikie/^ apparently a 

 strip of Arenig rocks has been wedged in against the Highland schists 

 along the great boundary fault with radiolarian cherts like those of the 

 south of Scotland and in the same sequence. The dull green diabasic 

 lavas show conspicuous ])illow and sack-like forms in the ravines of For- 

 farshire, and the igneous rocks are underlain, with cherts as in Ayrshire. 

 These rocks are about 50" miles from the nearest corresponding forma- 

 tions of the southern uplands, with which they may have been continuous, 

 as possibly also with the north of Ireland, where this series, probably 

 Arenig, attains its greatest development. 



From a review of the British examples, in 1911, Dewey and Flett^^ 

 concluded that ^*^the pillow lavas are a group of basic igneous rocks that 

 occur, in our experience, only as submarine flows.'^ The frequent asso- 

 ciation of cherts, many of them radiolarian, with this structure in Great 

 Britain, even where the flows are in coarse, shallow-water sediments, sug- 



^° A. Geikie : Ancient Volcanoes of Great Britain. London, 1897. p. 201. 

 =51 H. Dewey and .T. S. Flett : On some British pillow lavas and the rocks associated 

 with them. Geol. Mag., vol. viii, 1911, pp. 202-209, 241-248. 

 XLIV — Bull. Geol. See. Am., Vol. I25, 1913 



