Appendix: Two formal descriptions of the Bible's first eight words
1  Introduction
The set of numbers (or 'characteristic values') associated with the opening words of the Hebrew Scriptures may be represented as {G(i): 1 <= i <= 8}. Its members are found to be numerically related in two distinct ways, thus:
Here are the details:

G(i) = 37.p(i) + 6.q(i) i = word number G(i) = characteristic value P(i) =nearest multiple of 37 p(i)=multiplier Q(i)=excess/deficiency; a multiple of 6 q(i)=multiplier 

G(i) =100.r(i) + s(i) i = word number G(i) = characteristic value R(i) = nearest multiple of 100 r(i) = multiplying factor s(i) = excess/deficiency 
These relationships are completely independent of one another, and of all considerations discussed in earlier pages. Taken together, they reveal the extreme improbability of this sequence of numbers arising by chance.
2  Discussion
The coefficients in the foregoing relationships are interesting: 37 and 6 each a have high profile as a number per se, and are found together in both 37ashexagram and 37ashexagon; 100 is the square of 10, and thus has clear links with the human condition, and with metrication and decimalisation; 1  coefficient of s(i)  is the essence of number per se. Such coincidences  though not easily translated into 'odds against'  cannot be lightly discounted; in themselves, they confirm the unique status of their source.
However, from a theoretical standpoint, there is no problem assigning a probability to the matter of selecting a number at random to meet the requirements of these relationships. The following table relates to the first of these. It lists all numbers (N) in the range 0999 that meet the requirement: N = 37.p + 6.q :
p  q = 2  q = 1  q = 0  q = 1  q = 2  q = 3 
2  62  68  74  80  86  92 
3  99  105  111  117  123  129 
4  136  142  148  154  160  166 
5  173  179  185  191  197  203 
6  210  216  222  228  234  240 
7  247  253  259  265  271  277 
8  284  290  296  302  308  314 
9  321  327  333  339  345  351 
10  358  364  370  376  382  388 
11  395  401  407  413  419  425 
12  432  438  444  450  456  462 
13  469  475  481  487  493  499 
14  506  512  518  524  530  536 
15  543  549  555  561  567  573 
16  580  586  592  598  604  610 
17  617  623  629  635  641  647 
18  654  660  666  672  678  684 
19  691  697  703  709  715  721 
20  728  734  740  746  752  758 
21  765  771  777  783  789  795 
22  802  808  814  820  826  832 
23  839  845  851  857  863  869 
24  876  882  888  894  900  906 
25  913  919  925  931  937  943 
The table evaluations cover the range: 62<=N<=943  the number of values tabulated being 144 (the 8 values of interest highlighted). Clearly, the probability associated with the selection of a tabular value from the 882 values contained within this range is 144/882, or 0.163. It follows that the odds against finding an unbroken sequence of 7 after the first is 1/(0.163)^7, or 323,402 to 1!
Regarding the second relationship (completely independent of the first), it will be observed from the earlier table that the excesses/deficiences lie within a band of width 28 symmetrically disposed about the multiples of 100. The probability of selecting such a number at random is therefore 0.28, and the odds against selecting a sequence of 7 after the first is 1/(0.28)^7, or 7411 to 1.
Combining these probabilities, the odds against the opening sequence of 8 numbers being a product of chance exceeds 2 billion to 1. The order of magnitude of this figure is confirmed by reference to a copy of the preceding table in which those values failing to meet the requirements of the second relationship have been omitted.
p  q = 2  q = 1  q = 0  q = 1  q = 2  q = 3 
2  86  92  
3  99  105  111  
4  
5  191  197  203  
6  210  
7  
8  290  296  302  308  
9  
10  388  
11  395  401  407  413  
12  
13  487  493  499  
14  506  512  
15  
16  586  592  598  604  610  
17  
18  
19  691  697  703  709  
20  
21  789  795  
22  802  808  
23  
24  888  894  900  906  
25  913 
It is observed that only 41 values remain in the table  all meeting the required conditions. The probability of selecting any one of these from the total of 882 that lie within the range is therefore 41/882, or 0.0465, and of selecting an unbroken sequence of 7 after the first, (0.0465)^7. This is equivalent to odds against of 2.13 billion to 1.
A wideranging computer analysis of Hebrew text reveals that the language is favourably disposed toward the generation of characteristic values meeting the foregoing requirements. However this matter is interpreted, we are transported into the realm of the highly improbable when the independent geometrical data and apposite symbolisms (revealed in earlier pages) are included in the final reckoning.