roswell

to activate the ballast valve. A fixed ballast leak of about 400 graas
per hour was caused' by a defective valve fitting and this was suttialent
to maintain the balloon at nearly constant level until all the ballast was
exhausted. Following this experience, the use ot a preset fixed leak was
employed on many flights.
The very unstable "floating" seen on Flight 10, when the automatic
ballast valve controlled the flight. is not found on this flight Where the
vehicle used only a fixed-leak control. This eliminates both the over-
compensation and the serious effects ot temperature changes on the aneroid
capsule,. which are found when the automatic ballast valve is used.
The trajectory ot this balloon (Figure 8) shows a very interesting de-
formation at the transit ot the Sacramento Mountains. The anti-cyolonic
Figure 8
curvature over the eastern slope suggests that the air stream at the floating
level was distributed by the terrain. and the deformation predicted by d~c
theory may thus be given a physical illustration. The trajectory was de-
termined by aircraft and theodolite qbservation.
Another striking feature of the flight is the disagreement between the
actual flight path and the trajectory which might have been estimated tram
routine upper-wind reports. Reports from .l!il Paso, Roswell. Albuquerque and
White Sands were used for comparison with the observed trajectory.. Except
tor White Sands, none of these stations reported any wind from the WSW at or
near the floating level during the 12-hour period cqvered by the flight. At
White Sands a very shallow current was detected moving in the direction indicated
by the balloon flight. This clearly demonstrates the non-representiveness ot
the ord~nary pilot balloon observation.
-13-