This paper discusses the scale-dependent grasp.Suppose that a human approaches an object initially placed on atable and finally achieves an enveloping grasp. Under such initialand final conditions, he (or she) unconsciously changes the graspstrategy according to the size of objects, even though they havesimilar geometry. We call the grasp planning the scale-dependentgrasp. We find that grasp patterns are also changed according tothe surface friction and the geometry of cross section in additionto the scale of object. Focusing on column objects, we first classifythe grasp patterns and extract the essential motions so that we canconstruct grasp strategies applicable to multifingered robot hands.The grasp strategies constructed for robot hands are verified byexperiments. We also consider how a robot hand can recognizethe failure mode and how it can switch from one to another.