What are the psychological effects of relating to 3 person who features, among other things, instant compliance, unrelenting positive regard toward oneself, intrinsic charm, and no faults-in short, is 'perfect'? Many owners view their pets as this and more, yet obviously these characteristics exist mainly 'in the eye of the beholder' rather than being a function of the animal himself. Any person who tried to relate to real people as he did to his pet or judged real people relative to his pet would either have his expectations dashed or be sadly disappointed. Perfect fantasies have small correspondence with real life. Yet today in our society pet owners are encouraged to regard their fantasies (i.e., pets) as real, and to the degree that an ownec adopts this societal suggestion, psychologically the pet seems to function as a detriment to effective social relationships and consequently to the person's mental health. Recently, in harmony with such theoretic expectations, we found that pet owners claimed to like people as nor to feel as liked by others as non-owners ( 1 ) and our results were re-tested and extended in the present study. Three matched stratified random area samples of a country area, a town, and three major cities yielded 80, 306, and 122 Ss aged 12 to 89 yr. respectively. The self-report questionnaire was delivered by student volunteers; the rejection rate was less than 1 4 % . Ss filled out a rating scale consisting of four items: ( 1 ) how d o people-in-general like you?, ( 2 ) how d o you like people-in-general?, ( 3 ) how do you like petsin-general? and, if he owned a pet ( 4 ) how do you like your p e t ( s ) ? Ratings were made along a 10-unit line above which was printed very m u c h (superordinate to the numbers 1-2), considerably ( 3 4 ) , somewhat ( 5 6 ) , very little ( 7 8 ) , and not at all (9-10) . Test-retest reliabilities for the scales ranged from .6G to .94 over G wk. Half the sample took the Cameron Religious Dimensions scale and Canter's short form of the Barron Ego-Strength scale, while the rest were administered the Eysenck scale. For the town and city samples there were no age differences in claimed liking of people, while females and non-owners claimed to like people more; while the country sample evidenced no differences along these dimensions. Pet owners tended to claim less felt regard from others (3-way analysis of variance, F = 5.57; P < .07, one-tailed) as previously (Cameron, et al., 1966) noted. Urban pet owners scored lower on the Egostrength scale (3-way analysis of variance; F = 5.40; P < .03) while no difference o n the Eysenck neuroticisrn or extraversion scales was noted. Owners claimed that religion was of less value (F = 3.77; P < .06) as indexed by the Religious Dimensions scale. Owner's mean ranking of their liking of pets = 2.86 (SD = 1.74); of people = 3.17 (SD = 1.72); thus their claimed liking of pets exceeded their claimed liking of people ( t = 1.89; P < .07). Non-owner's mean ranking of their liking of people = 2.62 (SD = 1.37); of pets = 4.67 ( S D = 2.31); thus non-owner's claimed liking of people exceeded their claimed liking of pets ( t = 11.2; P < ,001) . Attachment to one's pet averaged 2.06, dearly within the very much affective area. While the overlap between pet owners and non-owners along each of these psychological domains was considerable, the differences suggest that pet owners are less psychologically healthy than non-owners.