Abstract. Let H denote the Friedrichs extension of a given semibounded operator H in a Hubert space. Assume XI < H, and X € a(H). If for a finite-dimensional projection P in the Hubert space we have I -P < Coast.(H -XI), then it follows that A is an eigenvalue of H, and the corresponding eigenspace is contained in the range of P. Using this, together with the known order structure on the family of selfadjoint extensions, with given lower bound 0, of minus the Laplace-Beltrami operator, we establish the identity Ug(\) =1 for all g B G for the following problem.U is a unitary representation of a Lie group G, and acts on the Hubert space L2(Q) for some Nikodym-domain 8cC.Moreover U is obtained as a certain normalized integral for the left-G-invariant vector fields on Í2, that is, for each such vector field X, the skew-adjoint operator dU(X) is an extension of X when regarded as a skew-symmetric operator in L2(ß) with domain Cg°(Q).1. Introduction. Recently, two results, related to the spectral theory of the Friedrichs extension, have appeared. In [7] it was shown that the SegalFuglede integration problem for constant coefficient partial differential operators [1] has an affirmative answer in a more general setting of Lie groups and symmetric spaces. The answer relates the geometry of a given finite-volume domain ß to the integration problem for Lie algebras of invariant vector fields on ñ. However, the spectral theory of the corresponding Laplace-Beltrami operators (and their extensions) is still not well understood. (Independently,in [6] it is established that the Friedrichs extension of a semibounded operator H has derivation properties with respect to a given operator algebra if it is known that the initial nonselfadjoint operator H has these properties.)The following theorem has implications for both of the above-mentioned problems, and it seems to be of independent interest (in view of the scarcity of general results on the point spectrum of the Friedrichs extension). The idea of the proof is based on that of Fuglede's main lemma [1], which is in fact about families of selfadjoint nonsemibounded operators. The results however have little in common.