The paper presents a study of the effect of well spacing on recoveries based on factual data from ten different fields. Data on comparative recoveries from the five Woodbine sand fault-line fields of Texas are cited as virtually providing a laboratory demonstration, in that other factors are constant but spacing varies through a relatively wide range. Since the fields are separate, these data can not be discredited on the grounds that the more closely spaced wells obtained their additional recoveries at the expense of the widely spaced ones.

Graphic charts incorporating a well-spacing scale in the form of an inverted diminishing series based on the expression 1/<rad>acres per well</rad> are used: (1) to show that the recoveries from the fault-line fields are almost precisely proportional to the reciprocals of the square roots of the respective well spacings expressed in terms of acres per well; (2) to show that the same relationship exists in cases where two different parts of the same field are developed on different spacing patterns; and (3) to study the deviation from that "normal expectancy" with intra-field groups where well density varies erratically within the field.

The hypothesis is advanced that the mathematical relation between well spacing and recoveries is an index of the inherent inefficiency of the well system of oil-field development.

It is also pointed out that arbitrary wide-spacing patterns may make it impossible to develop adequately accumulations against faults or small fields on sharply folded structures. The Crescent field, Oklahoma, is cited as a case in which fully half of the Ordovician production has come from a pay zone which would never have been tapped if the original 40-acre spacing pattern had been adhered to throughout the development of the field.

Attention is also called to the fact that widening the well spacing increases the ratio of dry holes, required to define a field, to producers in the field; and that the effect becomes more serious as field size diminishes.

In the discussion of the effect of well spacing on profits, it is pointed out that most such discussions in recent years have incorporated major fallacies which have served to "load" them heavily in favor of wide spacing. For instance, they have generally assumed that recovery is largely independent of spacing; consequently, they have concluded that the extra cost of additional wells over some arbitrarily selected spacing is all loss. They have also treated costs confined to the producing lease as if they represented the total cost of producing oil, completely ignoring the cost of finding and acquiring such productive leases.

An analysis of a hypothetical field is presented to show all the factors which must be taken into consideration to fairly judge the effect of various spacings on profits.

On the basis of the evidence presented, the following major conclusions are reached.

1. Well spacing does affect the recovery of oil from a reservoir; increased well density results in increased recovery; and within practical limits there is a definite mathematical relationship between the two.

2. Arbitrary wide-spacing regulations impose definite handicaps on the adequate exploitation of small accumulations on sharp folds or against faults, and increase the ratio of dry holes to development.

3. From the standpoint of cost per barrel of crude recovered or of profit per dollar expended, it is not invariably true that the wider of two spacings is the more economical. For a given set of conditions there is a certain spacing which is most economical; and either a closer or a wider spacing involves higher unit costs and lower unit profits.