It sounds fairly decisive.in contrast with [15] we do not observe

any systematic correlation between the HV, LV, IV, or

FV Hi maps and the CMB. Moreover, we do not observe

a significant correlation between Hi and the CMB in the

area of the sky defined by the Rectangular mask, as just

2 of the 89 maps find some correlation above the 99%

level, and for only one of the ℓ-ranges - consistent with a

chance occurrence.

We do, however, observe some correlation on degree

scales for a number of velocity slices, as seen in the third

row of Table I. The 15 maps that demonstrate a Kp2

masked V-band degree scale correlation above the 95%

level correspond to mean velocities of -405, -345, -325,

-315,-205, 105. . . 155, 215. . . 235, 345 km/s. Since this is

an a-posteriori observation and adjacent LAB slices are

highly correlated, it is impossible to ascertain the sta-

tistical significance of these. Nevertheless, we track the

correlation of 105. . . 155 km/s and 215. . . 235 km/s maps

to the same extended feature at (l, b) (−50◦,−45◦)

(none of the remaining 15 maps are significant above the

99% level). In Figure 3 we examine the correlation of 215

km/s with the different frequency bands to ascertain if

the signal is due to some kind of contamination from the

presence of this extended Hi feature. However, the signal

remains very consistent between the CMB bands indicat-

ing that there is no obvious foreground contamination on

these scales. The correlation also appears quite random

in nature,

Can someone explain the y-axes on the two graphs? They are slightly beyond my technical level - how do you divide by a beam?

Also, is 'a posteriori' really the right expression for 'deciding what to analyze after eyeballing the data first'?

*Bring back the crackpots!*