\relax 
\citation{ref1}
\citation{ref5}
\citation{ref6,ref7,ref8}
\citation{ref9}
\citation{ref6,ref10,ref11,ref12}
\citation{ref12}
\citation{ref13}
\citation{ref14}
\citation{ref15}
\citation{ref2}
\citation{ref16,ref17}
\citation{ref9,ref18}
\citation{ref19}
\citation{ref20,ref21}
\citation{ref22}
\citation{ref20,ref23}
\citation{ref24}
\@writefile{toc}{\contentsline {section}{\numberline {1}Introduction}{4}}
\citation{ref6}
\@writefile{lof}{\contentsline {figure}{\numberline {1}{\ignorespaces A diagram of the diffractive DIS process $ep \rightarrow eXp$ or $ep \rightarrow eXY$. The dotted line indicates where the diagram can be divided under the assumption of proton vertex factorisation. The kinematic variables are defined in section $2\hbox {}$.}}{5}}
\newlabel{fig:1}{{1}{5}}
\citation{ref14}
\citation{ref15,ref25}
\citation{ref5}
\@writefile{toc}{\contentsline {section}{\numberline {2}Kinematics and cross section definition}{6}}
\newlabel{sec:kinexsec}{{2}{6}}
\newlabel{eqn:xsec}{{3}{6}}
\newlabel{eqn:sigma}{{4}{6}}
\newlabel{eqn:myt}{{5}{6}}
\citation{ref6,ref10}
\citation{ref6}
\citation{ref6}
\citation{ref12}
\citation{ref3,ref4}
\citation{ref4,ref26}
\citation{ref27}
\citation{ref28}
\citation{ref29}
\citation{ref6}
\citation{ref4}
\@writefile{toc}{\contentsline {section}{\numberline {3}Models of {\relax \mathversion  {bold}{$F_L^D$}}}{7}}
\newlabel{models}{{3}{7}}
\newlabel{fld:nlo}{{6}{7}}
\citation{ref30}
\citation{cst}
\citation{Becker:2007ms}
\citation{ref31}
\@writefile{toc}{\contentsline {section}{\numberline {4}Experimental Method}{8}}
\@writefile{toc}{\contentsline {subsection}{\numberline {4.1}H1 detector}{8}}
\citation{ref6}
\citation{ref32}
\@writefile{toc}{\contentsline {subsection}{\numberline {4.2}Data samples }{9}}
\@writefile{lot}{\contentsline {table}{\numberline {1}{\ignorespaces Summary of the data samples used in the analysis.}}{9}}
\newlabel{tab:1}{{1}{9}}
\@writefile{toc}{\contentsline {subsection}{\numberline {4.3}Event selection}{9}}
\citation{ref33}
\citation{ref20}
\citation{ref20}
\citation{ref34}
\citation{ref35}
\citation{ref6}
\citation{ref37}
\citation{ref38}
\citation{ref39}
\citation{ref6}
\citation{ref36}
\citation{ref40}
\citation{ref41}
\citation{ref42}
\citation{ref43}
\@writefile{toc}{\contentsline {subsection}{\numberline {4.4}Corrections to the data and simulations}{11}}
\newlabel{subsec:datamc}{{4.4}{11}}
\citation{ref43}
\citation{ref6}
\@writefile{toc}{\contentsline {subsection}{\numberline {4.5}Cross section extraction}{12}}
\newlabel{subsec:xsec}{{4.5}{12}}
\citation{ref20}
\citation{ref20}
\@writefile{toc}{\contentsline {subsection}{\numberline {4.6}Systematic uncertainties}{13}}
\citation{ref43}
\citation{ref6}
\citation{ref20}
\@writefile{toc}{\contentsline {subsection}{\numberline {4.7}Extraction of {\relax \mathversion  {bold}$F_L^D$}}{15}}
\newlabel{subsec:FLD}{{4.7}{15}}
\citation{ref6}
\citation{ref4,ref26,ref27}
\citation{ref6}
\citation{ref29}
\citation{ref29}
\@writefile{toc}{\contentsline {subsection}{\numberline {4.8}Extraction of {\relax \mathversion  {bold}$R^D$} and the ratio {\relax \mathversion  {bold}$R^D / R$}}{16}}
\@writefile{toc}{\contentsline {section}{\numberline {5}Results}{16}}
\citation{ref6}
\citation{ref4,ref26}
\citation{ref44}
\bibcite{ref1}{1}
\@writefile{toc}{\contentsline {section}{\numberline {6}Conclusions}{18}}
\bibcite{ref5}{2}
\bibcite{ref6}{3}
\bibcite{ref7}{4}
\bibcite{ref8}{5}
\bibcite{ref9}{6}
\bibcite{ref10}{7}
\bibcite{ref11}{8}
\bibcite{ref12}{9}
\bibcite{ref13}{10}
\bibcite{ref14}{11}
\bibcite{ref15}{12}
\bibcite{ref2}{13}
\bibcite{ref16}{14}
\bibcite{ref17}{15}
\bibcite{ref18}{16}
\bibcite{ref19}{17}
\bibcite{ref20}{18}
\bibcite{ref21}{19}
\bibcite{ref22}{20}
\bibcite{ref23}{21}
\bibcite{ref24}{22}
\bibcite{ref25}{23}
\bibcite{ref4}{24}
\bibcite{ref3}{25}
\bibcite{ref26}{26}
\bibcite{ref27}{27}
\bibcite{ref28}{28}
\bibcite{ref29}{29}
\bibcite{ref30}{30}
\bibcite{cst}{31}
\bibcite{Becker:2007ms}{32}
\bibcite{ref31}{33}
\bibcite{ref32}{34}
\bibcite{ref33}{35}
\bibcite{ref34}{36}
\bibcite{ref35}{37}
\bibcite{ref37}{38}
\bibcite{ref38}{39}
\bibcite{ref39}{40}
\bibcite{ref36}{41}
\bibcite{ref40}{42}
\bibcite{ref41}{43}
\bibcite{ref42}{44}
\bibcite{ref43}{45}
\bibcite{ref44}{46}
\@writefile{lot}{\contentsline {table}{\numberline {2}{\ignorespaces  The diffractive reduced cross section $\sigma _r^D$ at $\sqrt  {s} = 225\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$, multiplied by $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$, measured with the $460\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ data, at fixed values of $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$, $Q^2$ and $\beta $. At the largest $\beta $, the bin-averaged cross section is given together with the lower and upper bin boundaries. The statistical ($\delta _{stat}$), uncorrelated ($\delta _{unc}$) and sum of all correlated ($\delta _{cor}$) uncertainties are given together with the total uncertainty ($\delta _{tot}$). The other columns show the individual correlated uncertainties, which are due to the positron energy scale ($\delta _{ele}$), the positron polar angle measurement ($\delta _{\theta }$), the LAr noise subtraction ($\delta _{noi}$), the hadronic SpaCal energy scale ($\delta _{spa}$), the efficiency of the logarithmic energy-weighted cluster radius cut ($\delta _{r_{log}}$), the charge asymmetry of the photoproduction background ($\delta _{asy}$), the model uncertainty due to the influence of $F_L^D$ ($\delta _{mod}$), the model uncertainty on the underlying $\beta $ and $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$ distributions ($\delta _{\beta }$, $\delta _{x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}}$), the influence of resonant ($\delta _{vm}$) and QED Compton ($\delta _{Com}$) contributions and finally the parametrisation choice for the bin centre corrections ($\delta _{bcc}$). A minus sign indicates that a source is anti-correlated with a change in the cross section. All uncertainties are are given in per cent. The normalisation uncertainty of $8.1\%$ is not included. }}{22}}
\newlabel{tab:2}{{2}{22}}
\@writefile{lot}{\contentsline {table}{\numberline {3}{\ignorespaces  The diffractive reduced cross section $\sigma _r^D$ at $\sqrt  {s} = 252\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$, multiplied by $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$, measured with the $575\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ data, at fixed values of $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$, $Q^2$ and $\beta $. At the largest $\beta $, the bin-averaged cross section is given together with the lower and upper bin boundaries. The description of columns $5-20$ is given in table $2\hbox {}$. All uncertainties are given in per cent. The normalisation uncertainty of $8.1\%$ is not included. }}{23}}
\newlabel{tab:3}{{3}{23}}
\@writefile{lot}{\contentsline {table}{\numberline {4}{\ignorespaces  The diffractive reduced cross section $\sigma _r^D$ at $\sqrt  {s} = 319\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$, multiplied by $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$, measured with the $920\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ data, at fixed values of $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$, $Q^2$ and $\beta $. At the largest $\beta $, the bin-averaged cross section is given together with the lower and upper bin boundaries. The description of columns $5-20$ is given in table $2\hbox {}$. All uncertainties are given in per cent. The normalisation uncertainty of $7.6\%$ is not included. }}{24}}
\newlabel{tab:4}{{4}{24}}
\@writefile{lot}{\contentsline {table}{\numberline {5}{\ignorespaces The diffractive structure functions $F_L^D$ and $F_2^D$ multiplied by $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$, at fixed values of $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$, $Q^2$ and $\beta $. The statistical uncertainty ($\delta _{stat}$), the sum of the statistical and uncorrelated uncertainties ($\delta _{stat+unc}$) and the sum of all correlated uncertainties ($\delta _{cor}$) are given together with the total uncertainty ($\delta _{tot}$). Absolute uncertainties are given. The values marked by an asterisk are upper limits at $95\%$ confidence level.}}{25}}
\newlabel{tab:5}{{5}{25}}
\@writefile{lot}{\contentsline {table}{\numberline {6}{\ignorespaces The ratio $R_D$ of the cross sections for longitudinally to transversely polarised photon cross sections, at fixed values of $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$, $Q^2$ and $\beta $. The sum of the statistical and uncorrelated uncertainties ($\delta _{stat+unc}$) and the sum of all correlated uncertainties ($\delta _{cor}$) are given together with the total uncertainty ($\delta _{tot}$). Absolute uncertainties are given. Data with $|R^D| > 50$ and a relative uncertainty larger than $100\%$ are not shown. }}{26}}
\newlabel{tab:6}{{6}{26}}
\citation{ref6}
\citation{ref6}
\@writefile{lof}{\contentsline {figure}{\numberline {2}{\ignorespaces  The energy distributions of the scattered positron candidates for the $460\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ (left) and $575\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ (right) data. The data shown as points are compared with the sum of the diffractive DIS MC simulation and background estimates (open histogram). The light-filled histogram shows the photoproduction background estimate from data, the dark-filled histogram is the sum of the QED Compton and inclusive DIS backgrounds, taken from MC simulations.}}{27}}
\newlabel{fig:2}{{2}{27}}
\@writefile{lof}{\contentsline {figure}{\numberline {3}{\ignorespaces The quantity $\Sigma _i(E-p_z)_i$ summed over all final state particles for the $460\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ (left) and $575\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ (right) data at high\nobreakspace  {}$y$. The data after background subtraction are shown as points, compared with the MC simulation shown as a histogram. The shaded area shows the effect of a variation of the hadronic SpaCal energy scale by its uncertainty of $5\%$.}}{27}}
\newlabel{fig:3}{{3}{27}}
\@writefile{lof}{\contentsline {figure}{\numberline {4}{\ignorespaces Distributions of the kinematic quantities $y$ (top), $\beta $ (middle) and $\qopname  \relax o{log}(x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P})$ (bottom) for the $460 \ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ (left), $575 \ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ (middle) and $920 \ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ (right) datasets. The data are shown as points compared with the sum of the MC simulation and background estimates (open histogram). The light-filled histogram shows the photoproduction background estimate from data, the dark-filled histogram is the sum of the QED Compton and inclusive DIS backgrounds, obtained from MC simulations.}}{28}}
\newlabel{fig:4}{{4}{28}}
\@writefile{lof}{\contentsline {figure}{\numberline {5}{\ignorespaces The diffractive reduced cross section $\sigma _r^D$ multiplied by $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$ as a function of $\beta $ at fixed $Q^2$ and $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$ for (from left to right) the $460\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$, $575\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$, $820\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ and $920\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ datasets. The data are compared with the predictions of H1 2006 DPDF Fit\nobreakspace  {}B (solid line), which is indicated as dotted beyond the range of validity of the fit. The dashed and dashed-dotted lines represent the contribution of $F_2^D$, which is the same for each beam energy. The inner error bars represent the statistical errors on the measurement, the outer error bars represent the statistical and total systematic uncertainties added in quadrature. The normalisation uncertainties of $7.6(8.1)\%$ for the $920(460,575)\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ data are not shown.}}{29}}
\newlabel{fig:5}{{5}{29}}
\@writefile{lof}{\contentsline {figure}{\numberline {6}{\ignorespaces The diffractive reduced cross section $\sigma _r^D$ multiplied by $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$ as a function of $y^2/Y_+$ at fixed $Q^2$, $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$ and $\beta $. The inner error bars represent the statistical uncertainties on the measurement, the outer error bars represent the statistical and total systematic uncertainties added in quadrature. The normalisation uncertainty is not shown. Up to four beam energies are shown, where the lowest $y^2/Y_+$ point is given by the $820\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ data for $Q^2 = 4\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}^2}$ and by the $920\ensuremath  {\@mathrm {\tmspace  +\thickmuskip {.2777em}GeV}}$ data at higher $Q^2$. The linear fits to the data are also shown as a solid line, the slope of which gives the value of $F_L^D$. The predictions and extrapolated predictions of H1 2006 DPDF Fit\nobreakspace  {}B are shown as dashed and dotted lines, respectively.}}{30}}
\newlabel{fig:6}{{6}{30}}
\@writefile{lof}{\contentsline {figure}{\numberline {7}{\ignorespaces The diffractive structure functions $F_L^D$ and $F_2^D$ multiplied by $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$ as a function of $\beta $ at fixed $Q^2$ and $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$. The $F_L^D$ data are shown as filled points, compared with the predictions of H1 2006 DPDF Fit\nobreakspace  {}A (dashed line), Fit\nobreakspace  {}B (solid line) and the Golec-Biernat and \L uszczak model (dashed and dotted line). The measurements of $F_2^D$ (open points) are compared with the prediction of H1 2006 DPDF Fit\nobreakspace  {}B (long dashed line). The inner error bars represent the statistical uncertainties on the measurement, the outer error bars represent the statistical and total systematic uncertainties added in quadrature. The normalisation uncertainty of $8.1\%$ is not shown. Upper limits on the value of $F_L^D$ at the $95\%$ confidence level in the highest $\beta $ bins are also shown.}}{31}}
\newlabel{fig:7}{{7}{31}}
\@writefile{lof}{\contentsline {figure}{\numberline {8}{\ignorespaces The diffractive longitudinal structure function $F_L^D$, divided by a parametrisation of the $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$ dependence of the reduced cross section $f_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P/p}$ \cite  {ref6}, as a function of $\beta $ at the indicated values of $Q^2$ and $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$. The data are compared with the predictions of H1 2006 DPDF Fit\nobreakspace  {}B (red line), which is indicated as dashed beyond the range of validity of the fit. The inner error bars represent the statistical uncertainties on the measurement, the outer error bars represent the statistical and total systematic uncertainties added in quadrature. The normalisation uncertainty of $8.1\%$ is not shown. The lowest of the two upper limits on the value of $F_L^D$ at the $95\%$ confidence level in the highest $\beta $ bin is also shown.}}{32}}
\newlabel{fig:8}{{8}{32}}
\@writefile{lof}{\contentsline {figure}{\numberline {9}{\ignorespaces  The ratio $R^D$ of cross sections for longitudinally to transversely polarised photons, as a function of $\beta $ at the indicated values of $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$ and $Q^2$. The data are compared with the predictions of H1 2006 DPDF Fit\nobreakspace  {}B, indicated as dashed beyond the range of validity of the fit. The inner error bars represent the statistical uncertainties on the measurement, the outer error bars represent the statistical and total systematic uncertainties added in quadrature. Data with $|R^D| > 50$ and a relative uncertainty larger than $100\%$ are not shown.}}{33}}
\newlabel{fig:9}{{9}{33}}
\@writefile{lof}{\contentsline {figure}{\numberline {10}{\ignorespaces  The ratio of $R^D/R$ as a function of $x$ at the indicated values of $Q^2$ and $x_{I\tmspace  -\thinmuskip {.1667em}\tmspace  -\thinmuskip {.1667em}P}$. The data are compared with the predicted ratio using H1 2006 DPDF Fit\nobreakspace  {}B / H1 PDF 2009 (solid line). The error bars represent the statistical and systematic uncertainties added in quadrature. Data with $|R^D/R| > 20$ and a relative uncertainty greater than $100\%$ are not shown. }}{34}}
\newlabel{fig:10}{{10}{34}}