""Our Lady of Martyrs," whether that castle was east or west of Aurie's Creek, evidences of Indian occupation having been found on a ihill on the west side of the creek as well as on a hill on the east side.^ These evidences, however, prove very little in determining the location of a particular castle three hundred years ago; they only become important when sustained by distances from given points or by natural features of record. The locative conclusion stated above is more positively emphasized by counting Van Curler's miles' travel and his landmarks in going west from Onekagoncka, and by the natural features which ■he noted in his Journal. Leaving Onekagoncka, he wrote that he walked "half a mile" (Dutch) "on the ice" which had frozen over the kill, or say one and one-half English miles, and in that distance passed "a village of six houses of the name of Canowarode." It was near the river obviously. Walking on the ice "another half mile" (Dutch), he passed "a village of twelve houses named Senatsycrossy." After walking "another mile or mile and a half" on the ice, he passed "great stretches of flat lands" and came to a castle w^ich he first called Medatshet, and later Canagere, which he denominated "The Second Castle." His distances traveling west "on the ice" were evidently more correctly computed than they were on his march on the rough path "along the kill that ran swiftly." His miles from Onekagoncka to Canagere are given as two and a half (Dutch) or about nine miles English. The actual distance is supposed to have been about eight. He found the castle "built on a hill without any palisades or any defence." He located it east -of Canajohare Creek, a stream which has never lost its identity. When Van Curler visited the castle it contained "sixteen houses, fifty, sixty, seventy or eighty paces long." Detained in this castle by a heavy fall of rain which broke up the streams -- the "January thaw" of 1635 in the Mohawk Valley -- Van Curler resumed his journey on the 20th, and "after marching