Having for a number of years read about Danehill's apparent liking for deep families, especially family 7, I thought a bit of research would be in order.

Anyway, the people who put forward the family 7 thing as a rational idea will never be drawn on how far back the 7-a strain can appear so I am going to use 5 generations as a cutoff for a family appearance.  This, of course, means any number of generations on the dam line. 

So for all Danehill offspring I scanned down the dam's 5th generation and found which families are in the good one's and also those in all Danehill offspring.  If family 7 is a major play then surely the ratio of good one's over all is going to be significantly higher than the other families found.  I found the following:

Family 7 means all members of family 7 including 7-a (which is the family most often put forward). I can almost feel people saying 'but it's 7-a that it is important' so at the bottom of the table I have shown family 7-a on it's own and the rest of family 7 on it's own.

Family 7-a does do better than the rest of family 7.  One could even note that if it was not for family 7-a then family 7 would be one of the worse families you could get!

Family 7-a does fine but there are even better whole groups of families like 6, 10. 12, 16 and 22.  There are also some very high scoring families like 24, 31, 52 and especially C2 and C26 but the numbers in these groups are quite small so care needs to be taken. 

I will leave people to make what they will of the numbers but to me there is no case here that supports the 'family 7 is the key' view.

Showing that family 7 appears more or less at the same good against bad ratio is one thing but it might be illuminating to look at how the all powerful 7-a potion arrived from the 5th generation into the mare being mated to Danehill for all the 26 group winning foals.

Of course, except for the dam tail all family 7-a occurences found in the 5th generation are going to be through stallions.  In the case of family 7-a we find some very good stallions.

Does this mean that the family 7-a thing is actually an affinity with a small number of top class stallions?

Does it make more sense to find a mare with Bletchingly or Rockefella or Pitcairn(we know they are top class stallions) close up rather than try to find a mare with Twine The Plaiden in the 16th generation?

But let's not stop here. What about those other families? Are they through just a few stallions as well?  The answer is that the vast majority of the families found in the better one's are found mainly through a small number of stallions.

Here is a breakdown of families found in Danehill's group winning progeny with the stallions through which those families are found (with number of times found in brackets):

Family 1 : Princequillo (53), Turn-To (45), Star Kingdom (33), My Babu (26), Court Martial (25), Klairon (22)
Family 2 : Northern Dancer (65), Sir Gaylord (47), Round Table (37), Aureole (19), Biscay (17)
Family 3 : Nashua (27), Forli (23)
Family 4 : Nearco (41), Hail To Reason (28), Ribot (16)
Family 5 : Native Dancer (51)
Family 6 : Sir Tristram (33) and Grey Sovereign (20)
Family 8 : Raise A Native (33), Sir Ivor (29), Bold Ruler (28), Nijinsky II (22), Relic (22)
Family 9 : Nasrullah (82), Tudor Minstrel (19)
Family 10 : To Market (11), Riverman (10)
Family 13 : Mr Prospector (18)
Family 14 : Nearctic (51)
Family 16 : Stardust (20), Crepello (17)
Family 19 : Never Bend (29)

Not a bad bunch are they? 

Danehill is of course dead so what mares he has worked with is irrelevant. However, he has got lots of stallions sons and what's good for the father might be good for the son so hopefully looking at what is going on with the female families is not a complete waste of time. 

To summarise we have two schools of thought as to which mares to use:
1) those with lots of Nasrullah, Northern Dancer, Princequillo etc
2) those with lines of Gibside Fairy foaled 1811.

If you are going to try option 2 then I would recommend using Bletchingly and Rockefella.  That's the lines that work and thankfully they are firmly in the option 1 category.

Adrian Parry Copyright 2006