Coition (he did not take up revolution at this point) differed from that due to other attractions. There are two and only two kinds of bodies that can attract: electric and magnetic.[125] Gilbert refined his position further by arguing that one does not even have magnetic attraction[126] but instead the mutual motion to union that he called coition.[127] In electric attraction, one has an action-passion relation of cause and effect with an external agent and a passive recipient; while in magnetic coition, both bodies act and are acted upon, and both move together.[128] Instead of an agent and a patient in coition,[129] one has "conactus." Coition, as the Latin origin of the term denoted, is always a concerted action. [130] This can be seen from the motions of two loadstones floating on water.[131] The mutual motion in coition was one of the reasons for Gilbert's rejection of the perpetual motion machine of Peregrinus.[132] [123] _Ibid._, ch. 1, pp. 45-46. [124] M: pp. 110, 314. [125] M: pp. 82, 105, 170, 172, 217. [126] M: p. 98. [127] M: pp. 100, 112, 113, 143, 148. It need hardly be pointed out that coitus is not an impersonal term. [128] M: p. 110. [129] M: p. 110. [130] M: pp. 109, 115, 148, 149, 155, 166, 174. [131] M: pp. 110, 155. [132] M: pp. 166, 332. See also footnote 84. Magnetic coition, unlike electric attraction, cannot be screened.[133] Hence it cannot be corporeal for it travels freely through bodies[134] and especially magnetic bodies;[135] one can understand the action of the armature on this basis.[136] Since coition cannot be prevented by shielding, it must have an immaterial cause.[137] [133] M: pp. 90, 106, 107, 108, 113, 132, 135, 136, 158. This is, of course, contrary to modern experience. [134] M: pp. 106, 107, 108, 114, 134, 136, 140, 162. [135] M: pp. 106, 109, 114, 159, 162. [136] M: pp. 137-140. [137] M: p. 109. Yet, unless one has the occult action-at-a-distance, change must be caused by contact forces. Gilbert resolved the paradox of combining contact forces with forces that cannot be shielded, by passing to a higher level of abstraction for the explanation of magnetic phenomena: he saw the contact as that of a form with matter. Although Gilbert remarked that the cause of magnetic phenomena did not fall within any of the categories of the formal causes of the Aristotelians, he did not renounce for this reason the medieval tradition. Actually there are many similarities between Gilbert's explanation of the loadstone's powers and that of St. Thomas. Magnetic coition is not due to any of the generic or specific forms of the Aristotelian elements, nor is it due to the primary qualities of any of their elements, nor is it due to the celestial "generans" of terrestrial change.[138]