1885 - Distribution of seats

[finsobruce]

Jun 20, 2022 14:56:44 GMT Pete Whitehead said:

Jun 20, 2022 14:51:29 GMT finsobruce said:

The old Liberal club was known as the Hatcham until its demise some time in the noughties. (picture of it in the link below).

I wonder when the name fell out of common usage. I have to say I've never come across it before.

whatpub.com/pubs/SEL/13577/hatcham-liberal-club-new-cross

Actually you have - and so have I. In fact the post above looks remarkably familiar vote-2012.proboards.com/post/1057125/thread

Gettin' old Pete, both of us.

St Catherines, Hatcham burnt down in May 1913 an attack blamed on suffragettes but not conclusively pinned on them as far as I can tell.

[islington]

Jun 20, 2022 14:56:44 GMT Pete Whitehead said:

Jun 20, 2022 14:51:29 GMT finsobruce said:

The old Liberal club was known as the Hatcham until its demise some time in the noughties. (picture of it in the link below).

I wonder when the name fell out of common usage. I have to say I've never come across it before.

whatpub.com/pubs/SEL/13577/hatcham-liberal-club-new-cross

Actually you have - and so have I. In fact the post above looks remarkably familiar vote-2012.proboards.com/post/1057125/thread

It's déjà vu all over again.

[islington]

Right, I've done some sums that I've been putting off for fear they might yield the wrong answer.

Assiduous readers of this thread will recall that the global figures for England, Wales, Scotland and Ireland are from the '1881 comparison' figures in the 1891 census, and likewise the global figures for counties and boroughs within each country.

For the individual countries and boroughs, however, some of the census figures seems hard to square with the approach known to have been taken in the 1885 redistribution; so I used the figures from the Debrett Parliamentary Guide. These show every sign of having been compiled with care, must have been derived from some official source, and were very similar to the census numbers, in most cases identical. They did, however, avoid the oddities in the census data, in particular not showing any boroughs below the politically agreed minimum of 15000.

However, I was conscious that I was using mixed sources and I worried that problems might emerge when I married them up.

I'm still to do Wales, Scotland and Ireland but I've now plucked up courage and summed all the English counties as given in Debrett. It comes to 12521742, slightly fewer (by 8110) than the census figure of 12529852. For English boroughs the Debrett figure is 12109396, a rather larger increase of 24233 compared with the census number of 12085163.

The crucial question is: does this affect the global apportionment between counties and boroughs? And the answer is ........ (wait for it) ...... NO. It reduces counties' PV for a 234th seat from 53661 to 53626, while boroughs' PV for 227 increases from 53356 to 53463. So it considerably narrows the gap but does not close it, and the split of England's 460 seats remains at 234-226 in favour of counties.

The Debrett numbers slightly increase the count for England as a whole from 24615015 to 24631138 but this is not enough to alter the apportionment between the various parts of the UK.

(Phew) Huge sighs of relief all round.

[islington]

I've now looked at other parts of the UK for discrepancies between census numbers and the figures I've been using.

I can find no issues at all in Ireland.

Edited to add: Wrong. There is a tiny discrepancy in Co Cork, for which the Debrett numbers add to 391109, 2 short of the census figure of 391111. This means that the Debrett figure for Ireland as a whole is 5174834, also 2 short of the census number. Needless to say, such a tiny discrepancy makes no difference to the overall apportionments.

In Wales, there are no problems with the boroughs but the Debrett figures for the counties sum to 922847, compared with the census figure of 920274. However, this slight increase of 2573 does not affect either apportionments within Wales or Wales's apportionment within the UK.

Scotland is another story. The census figures are hard to use because the numbers for Parliamentary counties and burghs in the 1881 census do not (obviously) take account of boundary changes in 1885, while in the 1891 census the relevant table, unlike the English/Welsh and Irish equivalents, does not include an '1881 comparison' column. So one must rely wholly on Debrett, and on checking my original notes I find I had failed to do this consistently. So I've gone back and applied Debrett numbers throughout Scotland and this produces significantly different figures in some cases.

The issue is not with Scotland as a whole, where the Debrett numbers sum to 3727848. This is short of the census number, 3728124, by a trivial 276 and such a small discrepancy does not, of course, affect Scotland's apportionment of 70 seats.

But when it comes to individual counties and boroughs, some of the changes are significant. Most obviously, several burghs, especially Glasgow, are larger, presumably reflecting boundary extensions in 1885, while a number of counties have correspondingly shrunk.

Counties now come out as follows.

Six
Lanarkshire - 360106

Two
Ayrshire - 161998
Aberdeenshire - 150129
Renfrewshire - 106832
Fife - 101751
Perthshire - 95044

One
Midlothian - 78901
Stirlingshire - 78000
Invernesshire - 72755
Ross & Cromarty - 72483
Angus - 67473
Argyll - 63506
Dunbartonshire - 61394
Orkney & Shetland - 57826
Dumfriesshire - 57575
Banffshire - 50875
Clackmannan & Kinross - 42057
Moray & Nairn - 38605
East Lothian - 38475
Wigtownshire - 38452
West Lothian - 37567
Roxburghshire - 37258
Berwickshire - 35379
Kirkcudbrightshire - 35054
Kincardine - 33349
Caithness - 30762
Sutherland - 22805
Peebles & Selkirk - 17966
Bute - 17489

And burghs thus.

Seven
Glasgow - 510202

Four
Edinburgh - 236032

Two
Dundee* - 140063
Aberdeen - 105003

One
Leith Burghs - 72856
Kilmarnock Burghs - 69618
Greenock - 65884
Montrose Burghs - 59674
Paisley - 55627
Falkirk Burghs - 53873
Ayr Burghs - 41723
Hawick Burghs - 37604
Stirling Burghs - 36780
Elgin Burghs - 31804
Kirkcaldy Burghs - 31424
Perth - 28949
Inverness Burghs - 26425
Dumfries Burghs - 25584
St Andrews Burghs - 19396
Wick Burghs - 17461

This outcome cannot be reconciled with a strictly mathematical approach. The underlying issue is that, although Scotland’s allowance of 70 seats accurately reflects its population (after allowing for the overrepresentation of Ireland and Wales), the country includes a large proportion of relatively small counties and burghs and these absorb so many seats that we are left with very few to represent adequately the larger units. Or to put it another way, with 59 seats preallocated according to the usual UK-wide rules, only 11 seats remain to be distributed according to population.

It seems that some kind of political or administrative settlement was imposed, and with only 11 seats in play one can see how this might have worked. In the first place, on the numbers it is hard to avoid giving 2 of the ‘spare’ seats to Edinburgh, raising its total to 4. This then leaves little alternative but to assign the remaining 9 to Glasgow and Lanarkshire. The arrangement adopted adds 5 to Glasgow (giving it 7 in all) and 4 to Lanarkshire (giving 6).

This is not a very satisfactory result: in particular, Glasgow has on the face of it a compelling case for an 8th seat (PV 68027). But this can be conferred only at the expense of Lanarkshire’s 6th (PV 65474) or Edinburgh’s 4th (PV 67438). But Lanarkshire is already a seat down on the 7 to which it would be entitled were it an English county; while Edinburgh would clearly receive 4 were it in England, and moreover may be thought due a degree of consideration on account of its status as the capital of Scotland. So, in view of the shortage of seats for distribution, one can see why the eventual outcome was arrived at. The shortage does not affect only Glasgow and (arguably) Lanarkshire - a county with Ayrshire’s population would surely have received a 3rd seat had it been anywhere else in the UK (PV 64799).

Glasgow 8's PV of 68027 was by some margin the largest in the UK not to be rewarded with a seat. It was followed at a considerable distance by Manchester 7 (65265) and Kensington 3 (65260). (In horse-racing parlance: "Won by a distance. A short head separated second and third.")

My conclusion is that a mathematical formula may well have been used to apportion seats between the nations of the UK, and was almost certainly used within England to distribute seats first between counties and boroughs collectively and then to individual counties and boroughs; but within Scotland and Ireland the exact distribution of seats, with respectively only 11 and 26 available after preallocation, was imposed as an administrative or political decision, albeit one that was very much guided by population. (I leave Wales out of consideration because only 3 seats were available after preallocation and whatever system was used it is difficult to see any possibility other than awarding them all to Glamorgan (as was indeed done).)   
 

[islington]

I feel we are now coming to the end of this exercise, and as I start to mop up, I'm generating some stats.

The 1885 redistribution completely eliminated the overrepresentation of boroughs compared with counties, which had been an entrenched feature of the system for centuries, especially in England, and which had been only partly and inconsistently addressed by the First and Second Reforms. The result was a dramatic rebalancing of representation within England: prior to 1885, boroughs returned 282 members compared with 172 from counties; after 1885, boroughs elected 226 and counties 234, strictly as per population.

Prior to 1832 each county or borough, no matter how large or small, had voted as an undifferentiated whole. The First Reform introduced, and the Second Reform extended, the practice of splitting counties into divisions; but 1885 greatly expanded this idea not only by providing that each county division, without exception, was to return a single member (most had previously returned two), but also by extending the practice to boroughs so that each of them, instead of returning its members (be they one, two, three, or in one case four) at large as previously, were in almost all cases split into the appropriate number of single-member divisions (with the exception of 24 two-member boroughs that continued to vote at large).

The result was dramatic. Going into the 1885 redistribution the House comprised 643 territorial members elected from 410 constituencies: 1 four-member, 12 three-member, 206 two-member and 191 one-member. After 1885 it comprised 661 territorial members from 637 constituencies: 24 two-member and 613 one-member. Ever since, the single-member seat has been the norm (although a few double seats lingered until 1950).

Someone upthread asked about distribution per county. Here are the numbers for the 40 traditional English counties (i.e. including Monmouthshire, and treating the IoW as part of Hampshire) including boroughs with the county (boroughs not part of any county, or straddling county boundaries, are assigned to the recognized parent county).

Bedfordshire: 3 (down from 4)

Berkshire: 5 (down from 8)

Buckinghamshire: 3 (down from 8)

Cambridgeshire: 4 (down from 5)

Cheshire: 13 (down from 14)

Cornwall: 7 (down from 13)

Cumberland: 6 (down from 8)

Derbyshire: 9 (up from 8)

Devon: 13 (down from 17)

Dorset: 4 (down from 10)

Durham: 16 (up from 13)

Essex: 11 (up from 10)

Gloucestershire: 11 (down from 13)

Hampshire: 12 (down from 16)

Herefordshire: 3 (down from 6)

Hertfordshire: 4 (no change)

Huntingdonshire: 2 (down from 3)

Kent: 19 (down from 21)

Lancashire: 57 (up from 32)

Leicestershire: 6 (no change)

Lincolnshire: 11 (down from 14)

Middlesex: 47 (up from 18)

Monmouthshire: 4 (up from 3)

Norfolk: 10 (no change)

Northamptonshire: 7 (down from 8)

Northumberland: 8 (down from 10)

Nottinghamshire: 7 (down from 10)

Oxfordshire: 4 (down from 7)

Rutland: 1 (down from 2)

Shropshire: 5 (down from 10)

Somerset: 10 (down from 11)

Staffordshire: 17 (down from 19)

Suffolk: 8 (down from 9)

Surrey: 22 (up from 11)

Sussex: 9 (down from 15)

Warwickshire: 14 (up from 11)

Westmorland: 2 (down from 3)

Wiltshire: 6 (down from 15)

Worcestershire: 8 (down from 11)

Yorkshire: 52 (up from 38)

Prior to 1885 Yorkshire had most MPs with 38. Its representation was increased to 52 but even so it yielded first place to Lancashire, up from 32 to 57. The largest increase, however, was 29 by Middlesex, up by 161% from 18 to 47. Surrey was the other double-figure increase, from 11 to 22. These four counties alone gained a staggering 79 seats, meaning that the two main northern industrial counties gained hugely in political influence, as did London. A further five counties had smaller increases, totalling 9 seats in all: Derbys, Durham, Essex, Mons, Warwks.

Three were unchanged: Herts, Leics, Norfolk.

The remaining 28 counties lost 82 MPs in total (bearing in mind that England's total territorial representation increased from 454 to 460). This averages just under 3 per county and the losses were fairly evenly spread - Wiltshire suffered most, going down from 15 to 6.

More stats to follow, I expect, but that will do for now.

[greatkingrat]

How would treating Monmouthshire under Wales change things? I'm guessing it probably wouldn't be entitled to the 4th seat in Wales.

[YL]

In the actual text of the Act, all of Wales is treated under England, not just Monmouthshire.

[islington]

Jul 2, 2022 16:20:13 GMT greatkingrat said:

How would treating Monmouthshire under Wales change things? I'm guessing it probably wouldn't be entitled to the 4th seat in Wales.

I don't think it would make much difference to Monmouthshire because its PV for a 3rd seat is 66094, which would qualify it easily in either England or Wales.

It would affect Glamorgan, though, because if Mons is included then Wales's pre-1885 allocation is 33 instead of 30, and this still leaves it overrepresented compared with the rest of the UK so it wouldn't gain any extra seats. So to generate that extra seat for Mons, something has to give somewhere, and it can only be Glamorgan 5 (PV only 52573). So Glamorgan would have to make do with 4, which to be honest wouldn't be a totally unreasonable outcome since now I come to check it (thanks for prompting me to look at this), Glamorgan 5 is the smallest PV anywhere in the UK to be rewarded with a seat.

This also suggests, incidentally, that the county/borough issue affected apportionments in Wales because Cardiff Boroughs' PV for a 2nd seat (which it did not get) is 57241, well ahead of Glam 5.

It's true, as YL says, that the Redistribution Act lumped Wales in with England, but it still must have been the case that Wales (without Mons) was treated separately for apportionment purposes because if it hadn't been, Glamorgan would have had to compete against English counties for its 5th seat and it wouldn't have qualified. In any case, there were references in Parliament to the importance of preserving Wales's  current allowance of 30, and this makes sense only if (a) Wales was seen as having something of a separate identity and (b) it didn't include Mons. 

[islington]

Jun 2, 2022 10:33:03 GMT YL said:

On that note, it's curious to see Cardiff's population, even combined with those of Llantrisant and Cowbridge (the other contributors to "Cardiff Boroughs"), not quite enough for a second seat. In 1918, with Penarth but without Llantrisant or Cowbridge and still without Llandaff, it had three seats.

Just to pick up on this, I hadn't realized to what extent Cardiff Boroughs was dominated by Cardiff itself. Assuming it contained the whole municipality of Cardiff, and I think it must have done, that accounts for 82761 of the total population of 85862, leaving only 3101 for the other two boroughs of Cowbridge and Llantrisant. I thought at first this must be wrong, but it turns out the population of Cowbridge MB was a mere 1229. I can't find a number for Llantrisant but that leaves 1872, which seems plausible given the very small number for Cowbridge.

Interestingly by 1891 the population for Cardiff alone leapt to 128915, apparently without any boundary extensions, so this was evidently a time of rapid growth and makes it more plausible that it should have received 3 seats in 1918.

For comparison, Swansea Boroughs was 100590 based on the 1881 census, of which Swansea itself contributed 76430, so the other boroughs (Aberavon, Kenfig, Loughor and Neath) were a more significant element. The fact that, being over 90000, Swansea Boroughs acquired a second seat in 1885 presented the boundary commission with the problem of how to split it into two divisions. Their solution can be seen below. Basically they drew a seat ('Town') comprising the main part of Swansea (this is what we call 'a nice sensible seat') but hived off northern parts of Swansea and put them in a seat ('District') with all the other contributory boroughs (this is what we call 'one of the strangest and least satisfactory seats ever devised'). The map shows Swansea alone - the other boroughs were scattered across the western half of Glamorgan.

If nothing else, however, this arrangement achieved close equality between the two seats: Town 50043, District 50547.

[YL]

I suppose a Swansea division and an "other boroughs" division would have been regarded as having an unacceptable discrepancy, even though that would have been what would have resulted had Swansea been defined as a borough in its own right with the other boroughs forming a "Neath Boroughs" (say) district. (Would Neath have been big enough to be a borough in its own right?) That poses a related question: how large discrepancies were tolerated when dividing a multi-member borough or county?

[islington]

Jul 3, 2022 15:11:14 GMT YL said:

I suppose a Swansea division and an "other boroughs" division would have been regarded as having an unacceptable discrepancy, even though that would have been what would have resulted had Swansea been defined as a borough in its own right with the other boroughs forming a "Neath Boroughs" (say) district. (Would Neath have been big enough to be a borough in its own right?) That poses a related question: how large discrepancies were tolerated when dividing a multi-member borough or county?

Trust YL to come up with a brief and apparently innocuous query that raises all sorts of ramifications.

Swansea Boroughs, sans Swansea, came to 24160 so it would have been a highly unbalanced split with the 'Swansea' division more than thrice the size of the 'Everything else' division.

It would have looked less bad if 'the rest' had been split off as a separate burghal district ('Neath Boroughs' isn't a bad suggestion) but although there were many surviving boroughs smaller than this, the smallest new borough created in 1885 was Fulham with 42900.

Anyway, I'm more and more gaining the impression that creating new boroughs or burghal districts wasn't within the boundary commissions' brief.

I still need to do some more research on this but my current working theory is that when the boundary commissions were appointed late in 1884 the Bill that would become the Redistribution Act 1885 had already been published and it had already taken many of the big decisions about which boroughs would be abolished, which would be aligned with the municipal boundaries, which new boroughs would be created in the MBW area and elsewhere, and so on. The boundary commissioners were then asked, firstly, to consider whether the boundary of a borough as currently existing or proposed in the Bill embraced “the whole of the population which ought to be included within the borough”. So this did not amount to the same type of comprehensive review of borough boundaries that had taken place in the First and Second Reforms, and it was not even the commissions' main job: it was more in the nature of a preliminary checking exercise to identify what (if any) additional changes to borough boundaries might be needed beyond those already in the Bill - perhaps, indeed, the English commission's recommendations were the 11 undivided boroughs that were mapped, plus about 4 divided boroughs that also had more complex changes.

Once this initial step was complete and the external boundaries were set, population tables could be prepared and seats apportioned to each county or borough accordingly; and finally the boundary commissions moved on to their main task and drew up proposed divisions of counties and boroughs.

This approach would be consistent with the fact that the commissions' instructions, at least as summarized in that impeccable academic source, Wikipedia, seem to give much more attention to how divisions should be drawn than to the setting of external boundaries.

To answer YL's final query, my impression on looking through the numbers is that within each borough or county the commissions tried to achieve reasonable equality between divisions. But in boroughs, they also had regard to factors such as wards and other existing boundaries, and the outcome varies from place to place: Liverpool's 9 seats, for instance, range from 52180 to 78569, while Manchester's 6 vary only between 67346 and 76217. And while I haven't done a proper analysis of this, the divisions in each county seem to cluster more closely around that county's average than is the case with boroughs. To take a couple of examples, Derbyshire's 7 seats are all between 53298 and 55514, and Co Durham's 8 between 58247 and 62700 (all numbers as per Debrett). Some regard seems to have been had to geographical factors - for instance, New Forest is noticeably larger than the other Hants seats, reflecting its relative isolation from the rest of the county; and Thanet is significantly smaller than the other seats in Kent.

The 90000 limit I think still applied. At a quick check (don't hold me to this), and looking strictly at divisions (as opposed to whole boroughs or counties) the largest I can find anywhere is the UK is South Ayrshire with 89256, but maybe that doesn't count because it is simply a carrying-forward unaltered of the existing split of Ayrshire as set in 1868 (I presume it would have been adjusted if it had exceeded 90000). Of divisions created for the first time in 1885, I think the highest population is Southwark (Bermondsey) with 84537; Marylebone (West) with 83871 and Glasgow (College) with 83665 seem to be the runners-up. The lowest are all in small counties divided into two: Huntingdon with 27412 is the smallest I can see, but this is a case where the whole county had only 57220 so it's not altogether surprising that the next smallest is the county's other division, Ramsey, with 29808.

These stats, of course, relate only to divisions. Once you throw undivided boroughs and counties into the mix, the largest constituency in the UK is Oldham (152513), but it of course has two members. Of single-member seats, the largest is St George Hanover Square with 89573 and the smallest Kilkenny with 15278. 

[islington]

A few further musings on the subject of divisions -

The largest ones tended naturally to occur in counties or boroughs that were near the top of the range for the number of seats awarded, so it's not surprising that Ayrshire provided the largest division in the UK. Ayrshire's population of 161998 was exceptionally large for 2 seats, especially for a county - it exceeded four counties in England and one in Ireland that received 3. It had been split into two single-member divisions in 1868 and it was a stroke of luck that the larger of these came in just below 90000 in 1881 so the existing boundary could be left alone.

The only larger 2-member unit was Kensington with 163151 and fortune smiled here too, because the obvious dividing-line along the Uxbridge Road (now Notting Hill Gate and Holland Park Avenue) yielded a very even split in terms of numbers: North 82517, South 80634.

The other large divisions mentioned in my previous post (Southwark (Bermondsey), Marylebone (West) and Glasgow (College)) all fell in boroughs that were near the top of the range for their respective numbers of seats.

Apart from those already mentioned, the only other instances I can see of divisions (not whole counties or boroughs) above 80000 are East Aberdeenshire 83295 and Tower Hamlets (Bow & Bromley) 82406. The former of these is, like South Ayrshire, a carry-over from 1868; as for the latter, as discussed upthread, Tower Hamlets was divided in a manner strongly influenced by the MBW districts and this means its 7 seats show considerable variation, perhaps worth setting out in full (smallest to largest) - Mile End 47491, St George 49382, Limehouse 56318, Stepney 58122, Whitechapel 71314, Poplar 74104, Bow & Bromley 82406.

Edited to add: The smallest borough division in the UK is apparently Paddington (North) with 45879 - this is a case of a borough on the small side (107218) for two seats, rather unevenly divided to fit with ward boundaries.

 

[John Chanin]

I’d just like to say that though this subject is outside my normal interests, I have found these discussions fascinating and would like to thank islington and YL for them.

[islington]

I am glad that others have found this thread of interest and I'd like to express my appreciation of other contributors.

As I move towards a conclusion, perhaps I should say a little about the 1885 redistribution in its historic context. I am talking here specifically about redistributions – that is, the creation, abolition or redrawing of Parliamentary seats – as opposed to the extension of the franchise.

This post is mainly new material but in order to keep the train of thought together there is an element of recycling from previous posts.

 

No one that has stayed with this thread will be surprised to learn that in my view, of all the Parliamentary redistributions in British political history, starting with the First Reform in 1832, that of 1885 is by far the most important. It applied a sweeping process of modernization to a system that was, notwithstanding admittedly important modifications in the First and Second Reforms, still in large part the ancient one of earlier centuries. Redistributions associated with previous Reforms had started with the status quo and then adjusted it to reflect other factors, notably population; but in 1885 these priorities were reversed and population became the primary factor, although other considerations, notably the status quo, were also taken into account. Moreover, the concept of the single-member seat, applied in a somewhat tentative way in the First and Second Reforms, was embraced wholeheartedly in 1885.

Another attractive feature of the 1885 redistribution is the way it was organized on a UK-wide basis, applying the same basic ground rules in all parts of the UK regarding the minimum size (15000) for any seat and the maximum number of population per MP (90000). Reflecting this approach, the detailed boundaries for the whole UK were enacted in a single piece of legislation, the Redistribution of Seats Act 1885.

The 1885 redistribution resulted in the greatest cull of boroughs in British political history. The 15000 limit was the chief factor here. Going into the 1885 redistribution, there were 98 boroughs below this level: 69 in England, 5 in Wales, 2 in Scotland, 22 in Ireland. Of these, one (Warwick) was rescued by the absorption of a larger adjacent town; another (Haverfordwest Boroughs) was merged into another burghal district. All the rest were abolished outright, including 7 (all in England) that had still returned two members. A further 8, all in England and all two-member, were abolished for other reasons even though they were above 15000: 2 for corruption, and the other 6 because, for historic reasons, they included large rural tracts and were not predominantly urban in character. So the redistribution accounted for 105 boroughs in all, completely exclipsing the 56 abolished in the First Reform. (NB: These figures do not take account of the reorganization of Parliamentary representation in the MBW area, which entailed the breaking up of 7 existing boroughs and the creation of 25 new ones.)

A further major change in the Third Reform, again representing a marked modernization of the system, was to lessen the importance of the age-old distinction between counties and boroughs. This was principally because it applied the same population rules to both but also because it brought about a closer alignment between the two in terms of the franchise, meaning it became less likely that a holding in a borough would qualify its owner to vote only for the county and not for the borough. Although the absence of a residential qualification for county voting remained an important distinction, the fact that county-only qualifications within boroughs were now relatively few meant that the overlap between the two could be disregarded and seats could be apportioned, their boundaries drawn, and their names bestowed, as if counties were wholly exclusive of boroughs in their area. This foreshadowed the Fourth Reform in 1918, when this distinction was made official.

It is notable that maps intended for popular consumption in the period 1885 to 1918 tend to be drawn in a way that implies there was no overlap between boroughs and counties; moreover, censuses of the period give the population of county divisions exclusive of boroughs.

The other major reason that the distinction became less important is that the overrepresentation of boroughs compared with counties, which had been an entrenched feature of the system for centuries, especially in England, and which had been only partly and inconsistently addressed by the First and Second Reforms, was completely eliminated in 1885. This meant a radical rebalancing of representation within England: prior to 1885, boroughs returned 282 members compared with 172 from counties; after 1885, boroughs elected 226 and counties 234.

Until 1832 each county or borough, no matter how large or small, had voted as an undifferentiated whole. The First Reform introduced, and the Second Reform extended, the practice of splitting counties into divisions; but 1885 greatly expanded this idea: not only by providing that each county with more than one member was to be split into divisions each of which, without exception, was to return a single member (most previous county divisions had returned two); but also by extending the practice to boroughs so that instead of, as previously, returning their members (be they one, two, three, or in one case four) at large, they were in almost all cases split, like counties, into the appropriate number of single-member divisions (with the exception of 24 two-member boroughs that continued to vote at large).

Going into the 1885 redistribution the House included 643 territorial members elected from 410 constituencies: 1 four-member, 12 three-member; 206 two-member and 191 one-member.

After 1885 there were 661 territorial members from 637 constituencies: 24 two-member and 613 one-member.

Prior to 1885 Yorkshire had most MPs with 38. Its representation was increased to 52 but even so it yielded first place to Lancashire, up from 32 to 57. The largest increase, however, was 29 by Middlesex, up by 161% from 18 to 47. Surrey was the other double-figure increase, from 11 to 22. These four counties alone gained a staggering 79 seats, meaning that the two main northern industrial counties gained hugely in political influence, as did London. A further five counties had smaller increases, totalling 9 seats in all: Derbys, Durham, Essex, Mons, Warwks.

Three were unchanged: Herts, Leics, Norfolk.

The remaining 28 counties lost 82 MPs in total (bearing in mind that England's total territorial representation increased from 454 to 460). This averages just under 3 per county and the losses were fairly evenly spread - Wiltshire suffered most, going down from 15 to 6.

The result was a decisive shift of political power away from rural areas and small country towns towards large towns and suburbs, especially London and the industrial north; a radical redistribution that far better fitted the country that Britain had become by 1885. And this was done in a way that generated a map far more familiar, to a modern eye, than the old arrangements it swept away.

In other words, the redistribution of 1885 was a pivotal point in the development of our representative machinery: the moment we moved decisively from the ancient system in favour of a recognizably modern approach. It is true that a few features of the older system stubbornly survived; but overall, 1885 set the tone for the future and it is not too much to say that each subsequent review, including the present one, is no more than an adjustment or refinement of the basic pattern set in 1885.

[J.G.Harston]

Jul 5, 2022 10:34:52 GMT islington said:

In other words, the redistribution of 1885 was a pivotal point in the development of our representative machinery: the moment we moved decisively from the ancient system in favour of a recognizably modern approach. It is true that a few features of the older system stubbornly survived; but overall, 1885 set the tone for the future and it is not too much to say that each subsequent review, including the present one, is no more than an adjustment or refinement of the basic pattern set in 1885.

I think 1885 also shows something we are good at, evolving a system rather than breaking it and building a new one. Even when the change is as big as 1885 it's within a process of manageable incremental change, not trying to do too much all at one (Sixth Review, I'm looking at you).

[minionofmidas]

Jul 5, 2022 11:54:02 GMT J.G.Harston said:

Jul 5, 2022 10:34:52 GMT islington said:

In other words, the redistribution of 1885 was a pivotal point in the development of our representative machinery: the moment we moved decisively from the ancient system in favour of a recognizably modern approach. It is true that a few features of the older system stubbornly survived; but overall, 1885 set the tone for the future and it is not too much to say that each subsequent review, including the present one, is no more than an adjustment or refinement of the basic pattern set in 1885.

I think 1885 also shows something we are good at, evolving a system rather than breaking it and building a new one. Even when the change is as big as 1885 it's within a process of manageable incremental change, not trying to do too much all at one (Sixth Review, I'm looking at you).

1885 did far more than the Cameronmander attempted to do - even if we ignore (as we should not; though passed separately and slightly later, the redistribution was intrinsically linked to it) the extension of the franchise.

The other difference is that this was a true compromise bill, agreed between Liberal and Conservative leadership (though also between the Tories and the Radicals, leaving the so far overrepresented Whig aristocracy out to dry.)

[islington]

Jul 5, 2022 13:25:04 GMT minionofmidas said:

Jul 5, 2022 11:54:02 GMT J.G.Harston said:

I think 1885 also shows something we are good at, (1) evolving a system rather than breaking it and building a new one. Even when the change is as big as 1885 it's within a process of manageable incremental change, not trying to do too much all at one (Sixth Review, I'm looking at you).

1885 did far more than the Cameronmander attempted to do - even if we ignore (as we should not; though passed separately and slightly later, the redistribution was intrinsically linked to it) (2) the extension of the franchise.

The other difference is that this was a true compromise bill, agreed between Liberal and Conservative leadership (though also between the Tories and the Radicals, (3) leaving the so far overrepresented Whig aristocracy out to dry.)

1. This comment is really true of the reform process as a whole. I look at it this way. All modern European states have Parliaments in one form or another; and if you dredge back into former times you will find that in the Mediaeval and Early Modern periods many of them also had some kind of national assembly under a variety of names. But I think I'm right in saying that in every case but one, certainly among larger states, the disappearance of the old assembly and the emergence of the new one are entirely separate processes: the modern Parliament was a new creation, not a continuation or revival of the old assembly. The splendid exception, of course, is the English / British Parliament, which has maintained an unbroken continuity from Mediaeval times right down to the present day; which means that to make it fit for the demands of the modern era, it had to undergo a major evolution. This is essentially what the Four Reforms were all about. As J.G.Harston says, "evolving a system rather than breaking it".
   
2. Indeed. The four great Reforms (1832, 1866-68, 1884-85 and 1918) all combined extending the franchise with redistributing Parliamentary seats. I suppose I agree that in the end the extension of the franchise is more important, but something I've found frustrating is that most histories of the period seem to concentrate almost entirely on the franchise aspect and their treatment of redistribution is often perfunctory. What I'm trying to do is redress the balance by focusing on the redistributive aspect. But I fully acknowledge the importance of the 1884 Act, which greatly extended the county franchise so that it aligned much more closely with the broader borough franchise that emerged from the Second Reform (and thereby among other things lessened the importance of the historic distinction between boroughs and counties). (I tend to refer to the 'First, Second, &c, Reform' rather than the 'First, &c, Reform Act' because each round of reforms involved more than one Act.)
3. Absolutely. But they were used to it, having been expertly played by Disraeli in 1866-68. So this was the second successive wave of reform in which they found themselves royally shafted, ironically after having had things very much their own way in 1832.
   

[islington]

A few more random stats re 1885 -

London was a big winner. If we define it as comprising the City and the MBW, going into the 1885 redistribution there were 10 boroughs largely or wholly in this area returning a total of 22 MPs. Afterwards there were 28 boroughs returning 59. If you omit the City and treat the MBW area alone, representation more than trebled from 18 to 57.

This is interesting because in mid-century, one of the main arguments against distribution according to population had been that this would result in an excessive allowance of seats to the London area. But by 1885, evidently, thinking had moved on so that this was now an acceptable outcome.

Of the 105 boroughs abolished in the UK (ignoring the complete reorganization of the MBW area), 76 were in England of which 61 were single-member boroughs. These included 34 of the 35 ancient boroughs that had been deprived of their second seat in 1868 under the Second Reform because their population at the 1861 census had been below 10000; the only survivor was Windsor, which had almost exactly doubled its population between 1861 and 1881 (from 9520 to 19082), partly because of a boundary extension in 1868 to include Eton and Clewer but also because of natural growth once the arrival of the railway brought it within easy commuting distance of London. 1885 also abolished 21 single-member ancient boroughs that had lost their second seat in 1832, and 3 (Abingdon, Banbury and Bewdley) that had always been single-member even before 1832. In addition, 3 single-member boroughs created in 1832 (Frome, Kendal and Whitby) had failed to grow as much as had (I presume) been anticipated in 1832 and were abolished because they were still below 15000 (Whitby at 14621 was perhaps unlucky).

15 two-member English boroughs were also abolished. 7 of these had been above 10000 in 1861 (thus had retained their second member in the Second Reform), but were still below 15000 in 1881: Barnstaple, Berwick-upon-Tweed, Newark, Tamworth, Tiverton, Truro, Weymouth. The other 8 were above 15000 in 1881 and were abolished for other reasons. Macclesfield and Sandwich were disfranchised for corruption and the others abolished because, for historic reasons, they were insufficiently urban in general character. Aylesbury, Cricklade, East Retford and Shoreham were ancient boroughs that had been punished for corruption in the decades before the First Reform by being 'thrown into the hundred', as it was called (i.e. the tiny original borough was massively expanded by the inclusion of a huge rural area so that its corrupt influence would be diluted as part of a much bigger seat). Wenlock had always been anomalous, its territory comprising 27 separate and largely rural pockets scattered across southern Shropshire; and Stroud had been made a borough in 1832 in the anticipation of industrial growth that never took place on the scale expected (someone predicted at the time it would become 'the Blackburn of Gloucestershire').

In Wales, Beaumaris Boroughs, Brecknock, Cardigan Boroughs, and Radnor Boroughs were all below 15000 and were abolished (Beaumaris and Cardigan were 14846 and 14751, so it shows they were serious about the 15000 rule and were not willing to allow special cases apart from Warwick). Haverfordwest Boroughs was also below 15000 and was merged into Pembroke Boroughs.

In Scotland only Haddington Burghs and Wigtown Burghs were below 15000 and both were abolished, although I should have thought there was a case for merging them with Hawick Burghs and Dumfries Burghs respectively.

Of Ireland's 31 boroughs going into 1885, 22 were abolished including really tiny places like Portarlington (2426 in 1881), Downpatrick (3902) and Dungannon (4081). Drogheda (14662) may have considered itself unlucky but, as I say, they were being strict about the 15000 rule.

The First Reform in 1832 is famous for abolishing 56 boroughs but the 1885 redistribution, although less well-known, easily eclipses this with 105: an essential step in creating a representative system fit for modern times.

  

[islington]

YL raised a query about how evenly counties and boroughs were split into their various divisions. This wasn't something I'd intended to look at but in trying to answer his question I unfortunately got interested and consequently have spent a considerable time, which I'll never get back (so, thanks a lot, YL ), investigating the distribution of borough and county divisions, alongside undivided boroughs and counties, within the permitted population range of 15000 to 90000 per MP. I divided the overall range into fifteen bands and the distribution of the 661 territorial seats comes out as follows. I've added the breakdown between the four UK nations, and England is further split between boroughs and counties).

15001 - 20000   023   E 15 (15/0); W 1; S 4; I 3

20001 - 25000   008   E 4 (3/1); W 3; S 1; I 0

25001 - 30000   024   E 18 (16/2); W 1; S 3; I 2

30001 - 35000   021   E 11 (9/2); W 2; S 4; I 4

35001 - 40000   028   E 9 (9/0); W 0; S 9; I 10

40001 - 45000   032   E 19 (14/5); W 3; S 2; I 8

45001 - 50000   094   E 68 (19/49); W 11; S 4; I 11

50001 - 55000   140   E 107 (17/90); W 4; S 6; I 23

55001 - 60000   111   E 78 (26/52); W 3; S 10; I 20

60001 - 65000   060   E 50 (28/22); W 0; S 5; I 5

65001 - 70000   060   E 42 (32/10); W 0; S 8; I 10

70001 - 75000   033   E 19 (18/1); W 1; S 8; I 5

75001 - 80000   012   E 9 (9/0); W 0; S 3; I 0

80001 - 85000   009   E 7 (7/0); W 0; S 2; I 0

85001 - 90000   006   E 4 (4/0); W 1; S 1; I 0

15001 - 90000   661   E 460 (226/234); W 30; S 70; I 101

There is a marked bunching in the middle. This is mainly because of the clustering of English county divisions, 191 of which (out of 234) fall in the range between 45001 and 60000.

Here are the 25 largest single-member seats:

89573 St George Hanover Square
89256 South Ayrshire
88128 Chelsea
87527 Dudley
87157 Huddersfield
85862 Cardiff Boroughs
84537 Southwark (Bermondsey)
84006 Birkenhead
83871 Marylebone (West)
83665 Glasgow (College)
83295 East Aberdeenshire
82517 Kensington (North)
82406 Tower Hamlets (Bow and Bromley)
80634 Kensington (South)
80036 Strand
79500 Glasgow (Bridgton)
78901 Midlothian
78840 Croydon
78739 Finsbury (Holborn)
78569 Liverpool (Everton)
78000 Stirlingshire
77890 Islington (East)
76217 Manchester (East)
76752 Deptford
75912 Hanley

And the 25 smallest:

21434 Rutland
21307 Rochester
19977 Stafford
19925 Montgomery Boroughs
19821 Hereford
19583 Bedford
19396 St Andrews Burghs
19295 Whitehaven
19171 Galway
19082 Windsor
18863 Boston
18539 King’s Lynn
18072 Penryn and Falmouth
17966 Peebles and Selkirk
17780 Winchester
17489 Bute
17461 Wick Burghs
17345 Grantham
16614 Taunton
16435 Salisbury
16111 Bury St Edmunds
15590 Newry
15372 Durham
15332 Pontefract
15278 Kilkenny

[Pete Whitehead]

This event might be of interest to some, especially islington www.history.ac.uk/events/voting-reform-150-years-1872-ballot-act