Without taking time and space to review what has been written on the development of IE velars and palatals + w in Greek and Italic, I submit the following examples in evidence of the usually accepted opinion that these do not develop alike. Those who hold that IE qw and k̂w fall together both in Greek and in Latin (cf. e. g. Persson Beitr. 124 ff., 520 ff.) virtually deny that there is any difference in the centum -languages between the pure velars and the palatals. Because IE qa - and k̂a - fall together in Greek and in Latin is no proof whatever that qwa - and k̂wa - would do the same. Such an inference is based on the erroneous assumption that IE qwa - became in early Greek xƑα-, in early Latin qwa-. That is begging the question, for then, of course, IE qwa - and k̂wa - would coincide in both languages.
A Wed, study studied this question.