In the first part of this paper, the author analyzes several unacceptable views of properties of physical objects (particulars) and justifies the familiar explication, according to which properties of physical objects are certain functions defined on the set of all couples w,t, where w is a possible world and t is a moment of time (it is assumed that the universe of discourse is the same at all couples w,t). On the background of the indicated understanding of properties, the author demonstrates that there exist empirical essential properties. An empirical property is understood as a property, the extension of which cannot be identified without applying experience. He distinguishes two kinds of essential properties: (i) purely essential and (ii) partially essential. A purely essential property has the same extension at every world-time couple. The extension of a partially essential properties changes in the logical space, but there exists a non-empty set of objects which is a subset of their extension at every world-time couple. Every partially essential property is empirical. The author pays special attention to the essential properties expressible by means of expressions of the form (λx)(P(x) ∨ Q(x)) where P, Q are non-essential empirical properties that are allied through some objects X1, ..., Xn - i.e., the property (λx)(P(x) ∨ Q(x)) belongs to the objects X1, ..., Xn at every world-time couple. In the last part of the paper, the author demonstrates that every empirical property of the type (λx)(x < I) where ”<” denotes the relation of being a part of, and ”I” stands for any physical object, is purely essential. The assumption that the extensions of these properties are changeable, entails unacceptable consequences.