524 PROCEEDINGS OF SECTION F. 
that the Greek Dionusos is derived from the Assyrian 
Daian-nisi, one of the names of the sun-god, or that Jason 
comes from the Greek verb “to heal,” I do not take upon 
myself to contradict him, but I have a right to ask, ““ Where 
are your proofs?”’ No proofs are forthcoming. These 
gentlemen do not deal in proofs; they only surmise. But 
they will have to deal in them nowadays if they want to be 
listened to. 
There must of necessity be myriads of surface resemblances 
in words without any connection at all. How many mil- 
lions of words are there in the thousands of languages now 
spoken on the face of the earth, to say nothing of the dead 
ones?) The human voice is capable of producing only a 
limited number of sounds; out of these few sounds, all the 
words spoken in all the earth have to be constructed, and 
the possibilities of their permutations and combinations are 
comparatively few. Is it not, then, a mathematical necessity 
that there should be numerous coincidences of phonology, 
and even, to some extent, of grammatical forms, without 
any linguistic connection at all? Surface-similarities in 
words, unsupported by corroborating evidence, are therefore 
positively of no value at all, excepting to suggest an inquiry 
as to whether there is anything in them. Science is becom- 
ing more exact, and more exacting, and some of our phil- 
ologists will have to come down from the clouds, and betake 
themselves to the practical methods of the natural-science 
raen, who are doing such splendid work by simply taking 
infinite pains, and recording nothing but well-ascertained 
facts. No slight work will be accepted nowadays, especially 
if it has a big theory hanging to it. 
