CoFFKV AND PiiAEGKK — T!ie Antrim llaiHcd Beach. 185 



site of a celt factory receutly discovered by Mr. Kiiowles at Cushen- 

 dall, County Antrim, are instructive on this point. The stone used was 

 chiefly basalt. The process of manufacture has been fully described 

 by Mr. Knowles.^ The final form aimed at was the sub-triangular 

 celt, the last stage of the chipping being carefully done, and serving 

 to trim the piece to the required thickness and shape, ready for 

 grinding and polishing. But in the early stages of the process the 

 final form is not directly attempted : in other words, the piece is not 

 prematurely specialised. At first it is freely chipped into an oval 

 form, pointed at both ends. The reason for this will be easily under- 

 stood when the nature of the work is considered. The block of stone 

 has first to be thinned down ; for that purpose it is flaked by striking 

 it along the margin of the upper face, flakes being thus detached from 

 the under face, the blow being directed downwards and inwards ; the 

 fracture which detaches tlie flake reaches in a considerable way on 

 the under surface. The block is then turned, and the process repeated 

 on the other face. In this manner the faces are thinned down and 

 the transverse section reduced to a flat oval. At this stage the 

 final shaping is begun, Tlie advantage of leaving the ends pointed 

 is now seen. The excess left at the ends allows the celt to be 

 thinned down lengthways at the butt and the cutting end while 

 bringing them to shape, in the same manner as the body of the 

 celt was thinned transversely by flaking from the sides. The finished 

 forms were rarely found at the quarry site on Tievebulliagh Mountain 

 where the stone was got, but only the roughed-out oval pieces, 

 many of them discarded pieces ; the final chipping was apparently 

 done at the valley sites, where the more finished forms were found. 

 The section of the rude pieces is often somewhat triangular, probably 

 intentionally in the case of adze-celt forms, but also a result of the 

 fact that the faces were flaked alternately, and that the flaking has 

 gone deeper on one face than on the other. 



It might be expected that we should find among these Larne 

 rejects, pieces which have been broken across in the ])rocess of 

 manufacture. AVe do occasionally find these roughed-out "Larne 

 celts " broken across. But as few sections of the gravels have been 

 dug out, and the majority of the specimens have been collected from 

 the distributed gravels of the Larne and Island Magee beaches, this 

 point has not been satisfactorily investigated. It may be pointed 

 out, however, that the Larne celt is stout in body, and not likely 



^ Joiini.i] of the Aiitbvopolonic'il Institute, vol. xxxiii., p. o60. 1903. 



