syntheses that weren't: w.f.r. weldon on chromosomes

Download Syntheses that Weren't: W.F.R. Weldon on Chromosomes

Post on 22-Jan-2018

18 views

Category:

Education

2 download

Embed Size (px)

TRANSCRIPT

  • Syntheses That Werent:

    W.F.R. Weldon on Chromosomes

    ishpssb 2017, University of So Paulo, 2017/06/18

    Charles H. Pence@pencechp

    Department of PhilosophyLSU Institute for Ethics

  • Outline

    1. Introduction and Background2. Weldon on Mendel3. Hereditary Transmission and Chromosomes4. e Problem5. From Chromosomes to Somatic Characters6. e Moral

    e take-home: Weldon worked toward, but nevercompleted, a combination of Galtonian inheritance,

    Mendels data, and a statistical view of the chromosomes

    Charles H. Pence 2 / 36

  • Introduction and

    Background

    Charles H. Pence Introduction 3 / 36

  • Weldons bust, Oxford Museum

    Charles H. Pence Introduction 4 / 36

  • Karl Pearson and Francis Galton

    Charles H. Pence Introduction 5 / 36

  • W.F.R. (Raphael) Weldon (18601906)

    Appointments: Cambridge (18821889); UniversityCollege London (18891899); Oxford (18991906)

    Early statistical work with Francis Galton, 18891890 Collaboration with Karl Pearson begins, 1890s Dispute with Bateson begins, 1894, escalates to openwarfare by early 1900s

    Rediscovery of Mendels work, 1900 Dies of acute pneumonia aged 46, 1906

    Charles H. Pence Introduction 6 / 36

  • Weldon on Mendel

    Charles H. Pence Weldon on Mendel 7 / 36

  • About pleasanter things, I have heard of and read a paperby one Mendel on the results of crossing peas, which Ithink you would like to read. It is in the Abhandlungendes Naturforschenden Vereines in Brnn for 1865 I havethe R.S. [Royal Society] copy here, but I will send it to youif you want it.

    e point seems to me to be that the results indicate anexclusive inheritance with a very high parental correlation.(PEARSON 11/1/22/40.4.2, 1900-10-16)

    Charles H. Pence Weldon on Mendel 8 / 36

  • I think there must be an element in each gametecorresponding to every quality transmitted by it; some ofthese may blend with the corresponding elements of theother, some may exclude corresponding elements of theother, some may make a patch work resulting in aparticulate inheritance. (PEARSON 11/1/22/40.4.3,1900-12-12)

    Charles H. Pence Weldon on Mendel 9 / 36

  • A Mean B C

    Charles H. Pence Weldon on Mendel 10 / 36

  • ere must be a simple relation between AB, BC, and theS.D. [standard deviation] of the original system, whichwould make the chance that a grandchild falls within BC =1/4?

    [A] more important point would be this: that since someraces of the same species are said to obey these laws, whileothers do not, one has a chance of predicting in a givencase, from a knowledge of parental variability, what willhappen, and so smashing the pure gamete once for all.(PEARSON 11/1/22/40.7.3, 1902-06-23)

    Charles H. Pence Weldon on Mendel 11 / 36

  • Hereditary

    Transmission and

    Chromosomes

    Charles H. Pence Chromosome Constitution 12 / 36

  • Dont worry over my attempts to understand inheritanceformulae ey will do me good, but otherwise they areof no importance, and you have plenty of better things todo. (WFRW to KP, PEARSON 11/1/22/40.6.4, 1901-12-31)

    Charles H. Pence Chromosome Constitution 13 / 36

  • n chromosomes,each p elts of R

    n chromosomes,each p elts of D

    2n chromosomes,n with p elts of Rn with p elts of D

    2n chromosomes,n with p elts of Rn with p elts of D

    F1 gamete F1 gamete

    F2 somatic cell

    Charles H. Pence Chromosome Constitution 14 / 36

  • RRR RRR

    DDD DDD

    RRRRRRGGGGGG

    merge all elements

    RRRor

    RRDor

    RDDor

    DDD

    RRRor

    RRDor

    RDDor

    DDDdiscard

    half of elements

    at random,

    form two new ch

    romosomes

    First Hypothesis: No individuality of the chromosomes

    Charles H. Pence Chromosome Constitution 15 / 36

  • RRR RRR

    DDD DDD

    RRRDDD

    RRRDDD

    merge paired chromosomes

    RRRor

    RRDor

    RDDor

    DDD

    RRRor

    RRDor

    RDDor

    DDDdiscard half of ele

    mentsat rand

    om

    Second Hypothesis: Individuality of the chromosomes

    Charles H. Pence Chromosome Constitution 16 / 36

  • The Problem

    Charles H. Pence The Problem 17 / 36

  • 10%

    20%

    30%

    12R 11R + D 10R + 2D 9R + 3D 8R + 4D 7R + 5D 6R + 6D 5R + 7D 4R + 8D 3R + 9D 2R + 10D R + 11D 12D

    n = 2, p = 3n = 3, p = 2

    no individuality

    Charles H. Pence The Problem 18 / 36

  • eyre all the same graph!

    Almost none of these will breed true.

    Charles H. Pence The Problem 19 / 36

  • I have laboriously worried through the eect of supposingthe chromosomes to retain their individual constitutionright on from the moment of fertilisation to the formationof new germ-cells; and it does not give anything like aproper segregation: there are very few pure individuals,among either dominants or recessives. (PEARSON11/1/22/40.11.1, 1904-01-11)

    Charles H. Pence The Problem 20 / 36

  • Chromosomes and

    Somatic Characters

    Charles H. Pence Chromosomes and Somatic Characters 21 / 36

  • Of course, the thing really wants more than this. It wantssegregation of elements, but what I have not and cannotput into it is the view of each element as a stirp. Dominance of yellow seed-colour in a Pea is not merelyprevalence of the eect, due to one set of elements, overthat due to the other set in the hybrid: it is also anexcitation of the ancestral yellow-seededness of the race ofgreen peas used in making the cross. (PEARSON11/1/22/40.11.1, 1904-01-11)

    Charles H. Pence Chromosomes and Somatic Characters 22 / 36

  • Here is the true gospel, or a sort of approximation to it, atlast! When a stirp goes into a zygote, it carries a lot ofproperties, but those which are manifested by the bodyinto which the zygote develops are transmitted withincreased intensity to the gametes of that body, thusestablishing that correlation between character of a parentand character of its reproductive cells, which I hadfoolishly been unable to put in.

    Charles H. Pence Chromosomes and Somatic Characters 23 / 36

  • But if a stirp, having become active in this way, beintroduced into a zygote in which the majority of stirpsare so active in other directions that its own propertiesbecome latent in the body to which the zygote gives rise,then that stirp transmits its properties in a weakenedcondition to the next generation. (PEARSON11/1/22/40.11.1, 1904-01-12)

    Charles H. Pence Chromosomes and Somatic Characters 24 / 36

  • I think this is good enough, at last! But I want to get thesegregation generations into the proper 1/4 + 1/2 + 1/4. To dothat, I must assume that the valency of a yellow stirpwhich has grown up in a hybrid body is (n/2 x) yellow= (n/2 + x) green where n is the whole number of stirps ina somatic element, = 2 whole number in a germ cell, andx is the probable error of the series representing thezygotes. [...] You see, I cant even nd the S. D. [standarddeviation], and if I could, I could certainly not nd therelation between it and the probable error! (PEARSON11/1/22/40.11.1, 1904-01-12)

    Charles H. Pence Chromosomes and Somatic Characters 25 / 36

  • I have been writing up a long account of the stirp, becauseI think it worth doing, in answer to Batesons statementthat the Law of Ancestral Heredity is incomplete, becauseit does not give a theory of the constitution of gametes.(PEARSON 11/1/22/40.11.2, 1904-04-17)

    We shall see in a future chapter how far the suggestionmade [by Galton] can be brought into harmony with thefacts of germinal structure which have been discoveredduring the last thirty years. (PEARSON 5/2/10/4, ch. 2,p. 2, f67r)

    Charles H. Pence Chromosomes and Somatic Characters 26 / 36

  • The Moral

    Charles H. Pence The Moral 27 / 36

  • Weldon had hoped that unaided statisticalconsiderations of the assortment of

    chromosomal elements would allow him toduplicate the Mendelian segregation patterns,along with the non-Mendelian patterns the

    biometricians saw in their data.

    Charles H. Pence The Moral 28 / 36

  • ey couldnt.

    Charles H. Pence The Moral 29 / 36

  • at meant that Weldon was forced to nd a wayto unite Galtons conception of patent and latent,active and inactive chromosomal elements with

    his statistical reasoning.

    Charles H. Pence The Moral 30 / 36

  • He never did.

    Charles H. Pence The Moral 31 / 36

  • But! We still learn several importantthings about biometry:

    Charles H. Pence The Moral 32 / 36

  • For Weldon, at least, this is why hewanted more data on plants, mice,horses, snails, and on and on. esehypotheses are hard to separate!

    Charles H. Pence The Moral 33 / 36

  • And the crucial next step for Weldon, athis death, was to determine the

    physiological, biological details thatsupported Galtons view of heredity. isis not data collection for its own sake.

    Charles H. Pence The Moral 34 / 36

  • Lastly, and perhaps most importantly,Weldons view of heredity unites several

    apparently incompatible ways ofunderstanding inheritance. We need allof Galtons picture, Mendels data, and

    contemporary cellular biology.

    Charles H. Pence The Moral 35 / 36

  • Questions?

    charles@charlespence.nethttp://charlespence.net

    @pencechp

  • It seems clear that such forms of segregation as I haveassumed (see Book MCMV) during nuclear division willnot lead to a separation of zygotes into classes sucientlysharp for Mendelian purposes, unless some persistence ofeect from one mitosis to another be assumed. eref

View more >