We present time-resolved spectrophotometry of the old nova BT Monocero
tis (Nova Monocerotis 1939). While BT Mon's light curve possibly sugge
sts the presence of an accretion disk, trailed spectrograms and Dopple
r tomograms of the data provide no such evidence. The spectra do revea
l, however, high-velocity gas motions near eclipse that are of a chara
cter unique among other cataclysmic variables studied similarly to dat
e. On the basis of the available data, we cannot describe the precise
nature of BT Mon, but we suggest and discuss four possible interpretat
ions of this extraordinary nova. Accretion by the system's white dwarf
via the standard stream-disk mechanism is roughly supported by the ap
pearance of BT Mon's light curve, yet our spectroscopy does not indica
te the presence of an accretion disk. And while accretion by the white
dwarf of a wind from the secondary star might explain the high-veloci
ty gas motions we observe surrounding eclipse, our spectroscopy provid
es no evidence for a wind. A third possible interpretation of our and
other's data for BT Mon involves two mass-transfer streams that emanat
e from the secondary star. One stream is the classical mass transfer s
tream; the other is a supersonic stream driven from the secondary star
by a de Laval nozzle. This two-stream model explains the appearance o
f BT Mon's unusual, high-velocity gas motions visible during eclipse,
but we find no additional evidence to support it. Finally, we consider
the possibility that the system's white dwarf is highly magnetized, a
ccretion occurring via an accretion column onto the primary star's sur
face. Our observations mildly support such an interpretation, and rece
nt results from other magnetic cataclysmic variables strengthen this l
ast interpretation.