Manorial surveys are the earliest development in manorial records, perhaps having pre-Conquest origins in the surveys of Tidenham and Hurstbourne and the Rectitudines Singularum Personarum. There is a sequence of manorial surveys, especially from the estates of religious houses, particularly large Benedictine houses of pre-Conquest foundation, from the late eleventh and through the twelfth century into the early thirteenth century.
During the thirteenth century, manorial surveys proliferated, and their production was continuous through the middle ages, in one form or another (see below).
At one level, survey can be accepted as a generic term comprehending slightly varying types of document including:
Although it is difficult fully to differentiate these types, the rental largely consists of the enumeration of tenants and the money rents owed, whilst custumal importantly defines (customary) services as well as lands, and the significance of the extent is that it gives an 'extended valuation'. In this last case, the 'extended valuation' consists in valuing every service [and the expected return on livestock and grain] in cash terms and producing at the end of the survey an annually expected return from the manor. The development of the extent is considered further below.
The comparison between twelfth- and thirteenth-century surveys has been employed by some historians to propose a radical change in both estate policies and the status of the unfree peasantry. Initially in his article on the chronology of labour services and then in subsequent work, Postan suggested the disintegration of the demesne in the twelfth century. Noticing the appearance of apparently new impositions on the unfree peasantry, Hilton proposed a decline in the status of the unfree between the middle of the twelfth and the middle of the thirteenth century.
In the late twelfth century and early thirteenth century, demesnes were resumed in hand by lords and exploited directly by them rather than leasing out. This change of practice did introduce an interesting change in the content of the surveys. Whilst demesnes were leased out in much of the twelfth century, the arrangement was predominantly made under stock and land leases, by which the lessee received not only the land and the incumbent services, but also the demesne stock and buildings. The stock and buildings were to be returned in the same number and condition when the lease expired. Consequently, the stock, and sometimes the buildings, is listed in the survey, usually near the description of the demesnes as one of the first items in the survey. When demesnes were resumed in hand, it was no longer necessary to record the stock and land in the surveys, so these items disappeared from the survey, although the description of the demesne remained. Henceforth, the stock was only accounted for in the manorial accounts.
Although thirteenth-century surveys enumerate services in considerable detail, the obligations of the peasantry, especially the unfree, should be assumed to be theoretical and not necessarily exacted at the time the survey was compiled. In fact, some of the customary services may have been sold or commuted, which is only known from manorial accounts. For the services, surveys are a theoretical statement of the obligations. On the other hand, they may well reflect on conditions at an earlier time. For example, although customary carrying services enumerated in the thirteenth-century surveys may have been sold at the time the survey was compiled, they may have been used in the twelfth century and the details recorded in thirteenth-century surveys may thus reveal much about carrying services in the twelfth century.
Quality and quantity of services: the quantity of the services was defined quite precisely in the thirteenth-century surveys; what distinguished the villein, at least theoretically, was the quality as well as the quantity; the unfree did not know what they would have to do from one day to the next, since it was a matter of the lord's discretion; in fact, there was a tendency for this to be defined too.
Comprehensive? Survey-type documents are basically concerned to enumerate tenants, not the entire manorial population. Rarely do surveys include sub-tenants, under-tenants or undersettles, although a few did (Caddington, Beds, in 1297 and Havering, Essex, in 1251/2).
The extent developed in the thirteenth century, perhaps influenced by the extended valuations introduced in Inquisitiones post mortem from the 1260s. Its rationale was formalized in the Extenta Manerii of 1276. One rationale of the extent was to compare the extent with the manorial account to assess how well demesnes were producing, one form of the 'profit-and-loss' calculation in the late thirteenth century, of which the best example is the Red Book of Worcester of 1288.
The extract is for the manor of Somersham which comprised several berewicks; what is provided here is the main entry, for Somersham alone. The structure is as follows:
Note that the demesne is described in parcels of acres, so that there are lots (I mean, lots) of numbers (e.g. quadraginta, tresdecim, quatuordecim); note too that there are half acres and that, since [nom sing] acra is a feminine noun, the adjective dimidius/dimidia/dimidium must be declined to agree with acra. There are also roods (nom sing roda, nom pl rode).
Note that the characters thorn /þ/ and eth /ð/ occur.
 gurges (gurgites).
fascula virgarum [from virga].
 custodiat (from custodire).
metentes (reaping -- from metere); metentem later (acc sing masc/fem).
precaria (boonwork; here precariam -- acc sing)
Episcopi -- heavily contracted -- 'of the Bishop'; next line Episcopo
inuenire occurs frequently
tallagiari (to be tallaged -- from tallagiare to tallage) (here as tallag')
[3 Customary tenants] operabilem (declined with virgatam here; nom fem = operabilis)
ebdomada (week; sometimes septimana)
 operacio(nes) (usually just opus/opera)
a die nathalis usque ad Epiphaniam
pannagio (from pannagium)
preter unam suem (except for a sow)
gallina/galline (nom) (here gallinas)
oua (from ovum (egg -- not to be confused with ovis/oves ewe/ewes (but sometimes just sheep, although bidentes is more usual for sheep -- ewes are usually more fully oves matrices from ovis matrix)
aueragium (carrying service)
secundum turnum vicinorum
socii (from socius)
aduentu (abl) (visit)
triturare (to thresh)
auena (oats -- here auene, of oats)
ad prebendam (for fodder)
braseum (malt) (here brasei, of malt)
sarclandum (from sarclare, sometimes sarculare, to weed, hoe)
frumentum (wheat -- here frumenti of wheat)
arabit (from arare, to plough)
in hieme (in winter)
in quadragesima (in Lent)
caruca (a plough -- here carucam)
cum denariis suis (with his own money -- i.e. at his own cost)
Cottars  super dorsum suum (on his back); fenum (hay); in crastino (on the morrow, from crastinum))
That completes most of the vocabulary of this survey.