[en] It is conjectured that euclidean geometry should be derived from spinor geometry through the equivalence of simple semispinor with isotropic semi n-vectors. The only tensors of complex 2n dimensional Euclidean space Esub(c)sup(2n) should then be: isotropic n - vectors and their intersections. Esub(c)₄ spinor geometry generates two isotropic semi bivectors equivalent to the semispinors of Esub(c)₄ (their geometrical properties are those of light propagating in vacuum), and their intersection: an isotropic vector (possibly representing momenta of massless particle and/or light rays); but no scalar, pseudoscalar or pseudovector is generated. In order to generate vectors outside the light cone in Msup(3.1) one needs not less than Esub(c)₆ spinor geometry which also generates Lorentz pseudoscalars and non isotropic pseudovectors and tensors. Besides, Dirac spinor should then always appear in doublets in Msup(3.1). Furthermore the mere geometrical structure of Esub(c)₆ spinor geometry seems to suggest formally, both Poincare (extended) and conformal supersymmetry. The suggested spinor-geometrical approach privileges the elementary role of semispinors. Its relevance for the real world should be manifested by the privileged role of semispinors in elementary interactions as in fact seems to be the case with Lorentz semispinors in weak interactions (and could perhaps also be the case for strong ones where conformal semispinors (or twistors) could be the interacting spinor fields). (author)