of block diagonality of the inverse covariance matrices W and W, for when 
these matrices are of the respective forms 
W dine M eer e) 
(6m, 6m) (618) (626 (6,6) 
(2) P2281 s 
diag ( W, Wut tam 
W = 
the general system of normal equations assumes the following fine structure 
  
  
  
  
  
  
° . je E e e e 7 
NSW 0 D 0 } af Qo Nia | | & q -Me 
e * 1 e e e 
o Un We e. 0 N21 N22... Nan | | 82 €? = Wr 
3 
. e. €. i . *. e . e 
e e e i e e e e e. 
° ° . Ü e ° e ° ° 
© ° 4 . e e eo 
i sen — -— 
0 0 ... N x + W a : Na Nm, seo N 6 c = We 
€» GN «p am GP GU Ge Sv Ov Go Bh GN m Gà QD CD €» OU G9 SE EE Ge Sm We OW 3 e G mo us qe Gm U ae ee Go BY GY oe © Gn ou GO €» €» SN Ge UD u. ee = = = e» s € €» Ww We 99 
(3) —-— T -— T oT 1 LI] LI . .. ee es . 
Ni Na s 0 Na IN*W D ees 0 & q-W 6 
dem T i T b T 1 e ee oe oe eo e 
Ni? N22 ... No § 0 N, + Ww, ... 0 52 C2 - W»€» 
i 
e ° e. 8 e e e e e 
e. . 9 i e. e. r] e. e. 
° e 8 e e e e e. 
8 : 
Kl M N 1 0 0 ec oe t vo ee oo 
In 2n ev mn 8 ses NW. n | | ^n - Wi n 
„A — . a) 
in which 
à; = 6x1 vector of corrections to elements of exterior orien- 
tation of Zth photo, 
Su = 3x1 vector of corrections to coordinates of jth object point, 
£, = 6x1 subvector of € corresponding to 1th photo, 
£j = 3x1 subvector of É corresponding to jth point, 
and 
e n . e n e 
N; = D N; j ’ CP p ej : 
J=1 J=1 
(4) 
Hg eo ® oglips Bd beer pu Ho fapen Bd 
The terms NN Nr 20s all appear in the following indeterminate sub- 
system of normal equations 
CO 
1i 
im] 
as 
sm. 
le: 
fot 
ap 
aer 
in 
nor 
It 
of 
nor 
AP| 
the 
red 
the