364 THE CANADIAN NATURALIST. [Vol. ix. 



Sumerians with their Khita confederates. They left their 

 traces in Media, as at Ujan, where druidical circles are found 

 which "M. D. Hancarville regarded as resembling and probably 

 coeval with the stupendous British monument Stonehenge." Dr. 

 Ferguson, in his " Rude Stone Monuments in all Countries," 

 finds these stonehenges in Northern Africa, Asia Minor, and even 

 in India, and maintains their common origin. In Peru we have 

 found them as the work of a Sumerian people ; and I am firmly 

 convinced that, wherever else they may be discovered, such, philo- 

 logy, coming to the aid of archaeology, will show to have been 

 their orisrin. 



It may appear a somewhat improbable thing that a people 

 speaking a Semitic language, such as was the ancient Himyaritic, 

 should connect with the so-called Aryan Cymii and other Celts. 

 Sir Gardner Wilkinson, however, speaking of that sub-Semitic 

 language, the Egyptian, states that it has affinities with the 

 Celtic and the languages of Africa, and adds : " Dr, Ch. Meyer 

 thinks that Celtic in all its non-Sanscritic features most strikingly 

 corresponds with the old Egyptian." We have already seen 

 that the Berber and Hausga, both in point of grammar and voca- 

 bulary, present much in common with the Celtic, and that there 

 are well defined Celtic traces in the Accad arising from the 

 Sumerian relations of that language. The Sumerian seems to 

 have been from the beginning a language peculiarly susceptible 

 of surroundino- influences, so that, while in Arabia and Africa 

 it retained a Semitic character, in Europe it approached the 

 Aryan, and in Chaldea and Peru became thoroughly Turanian. 

 The Celtic dialects contain a great many Semitic roots, in the 

 the possession of which they differ entirely from the Indo-Euro- 

 pean languages, as they also differ from them, while agreeing 

 with the Semitic tongues, in several grammatical forms of no 

 small importance. 



The occurrence of megalithic structures, so much resembling 

 the Cymric erection called Stonehenge as to call for comparison 

 with that monument from many different writers, in constant 

 connection with Sumerian forms, is an argument that applies to 

 the Arabian Himyarites as well as to other peoples whose lan- 

 o'uao-e ao-rees better with the Celtic. Stonehenge itself was 

 known as the work of Emrys and was a Cymric structure ; those 

 erections of a similar nature, referred to by Ferguson, Cather- 

 wood and Pegot-Ogier, as found in Northern Africa, relate to the 



