Visual inspection of stars embedded in H II nebulae has shown a signif
icant fraction to be surrounded by nearly symmetric extended regions w
ithin which the nebular brightness is apparently significantly fainter
than is typical for the surrounding area. These ''socket stars'' migh
t be caused by a bubble in the nebula blown out by a stellar wind or t
hey might be caused by a circumstellar envelope of dust hiding the emi
ssion behind the star. As such, the sockets could be the first manifes
tation of a previously unknown component of pre-main-sequence stars. U
nfortunately, no quantitative proof of the existence of sockets has be
en presented. To fill this need, I have imaged 10 socket stars and six
background stars with CCD cameras and infrared array cameras. From th
ese images, I have constructed radial plots which should reveal dips i
n brightness immediately outside the seeing disk. The radial plots do
not show any evidence for the existence of sockets. A detailed examina
tion of the photographs originally used to identify the sockets shows
that the causes of these reports are (1) artifacts resulting from the
photographic process of dodging and (2) random coincidence of stars wi
th local minima in nebular brightness. Thus, I conclude that ''socket
stars'' do not exist.