1: \documentclass[a4paper]{mem}
2: \usepackage{natbib}
3: \usepackage{graphicx}
4: \usepackage[a4paper]{hyperref}
5: \idline{73}{1}
6: %\documentstyle[aas2pp4,twoside,natbib]{article}
7: %\documentstyle[aasms4,twoside,aabib]{article}
8: %\setlength{\oddsidemargin}{0.2cm}
9: %\setlength{\evensidemargin}{-1.0cm}
10: \input epsf
11: %
12: \def\apj{ApJ}
13: \def\apjs{ApJS}
14: \def\apss{Ap\&SS}
15: \def\aj{AJ}
16: \def\apjl{ApJL}
17: \def\pasj{Proc. Astron. Soc. Japan}
18: \def\prl{Phys. Rev. Lett.}
19: \def\pre{Phys. Rev. E}
20: \def\aap{A\&A}
21: \def\araa{ARA\&A}
22: \def\mnras{MNRAS}
23: \def\jcp{J. Comp. Phys.}
24: \def\pasp{PASP}
25: \newcommand{\rmn}[1]{\mathrm {#1}}
26: %
27: \def\s2n{S^{\prime}/N}
28: \def\vecr{{\bf r}}
29: \def\vecdr{{\bf \Delta r}}
30: \def\dr{{\Delta r}}
31: \def\aq{{\langle Q \rangle}}
32: \def\jco{{J=1-0 $^{13}$CO}}
33: \def\co{$^{13}$CO}
34: %
35:
36: \bibpunct{(}{)}{;}{a}{}{,}
37:
38: % $Id: ppadoan.tex,v 1.1 2004/09/27 05:37:33 padoan Exp $
39:
40: \begin{document}
41: \title{Brown Dwarfs from Turbulent Fragmentation}
42:
43: \author{
44: Paolo Padoan\inst{1}, Alexei Kritsuk\inst{1}, Michael, L. Norman\inst{1} \and \AA ke Nordlund\inst{2}\fnmsep
45: }
46: \institute{Department of Physics and Center for Astrophysics and Space Sciences,
47: University of California, San Diego, 9500 Gilman Drive, La Jolla, CA 92093-0424; ppadoan@ucsd.edu, akritsuk@cosmos.ucsd.edu, mnorman@cosmos.ucsd.edu \\
48: \and Astronomical Observatory / NBIfAFG, Juliane Maries Vej 30, DK-2100, Copenhagen, Denmark; aake@astro.ku.dk
49: }
50:
51:
52: \abstract{
53: The origin of brown dwarfs (BDs) is an important component of the
54: theory of star formation, because BDs are approximately as numerous
55: as solar mass stars. It has been suggested that BDs originate from
56: the gravitational fragmentation of protostellar disks, a very different
57: mechanism from the formation of hydrogen burning stars. We propose
58: that BDs are instead formed by the process of turbulent fragmentation,
59: like more massive stars. In numerical simulations of turbulence and
60: star formation we find that gravitationally unstable density peaks
61: of BD mass are commonly formed by the turbulent flow. These density
62: peaks collapse into BD mass objects with circumstellar disks, like
63: more massive protostars. We rely on numerical experiments with very
64: large resolution, achieved with adaptive mesh refinement (AMR). The
65: turbulence simulation presented here is the first AMR turbulence
66: experiment ever attempted and achieves an effective resolution of
67: $1024^3$ computational zones. The star formation simulation
68: achieves an effective resolution of $(10^6)^3$ computational zones,
69: from a cloud size of 5~pc to protostellar disks resolved down to 1~AU.
70: }
71: \authorrunning{Padoan et al.}
72: \titlerunning{Brown Dwarfs from Turbulent Fragmentation}
73: \maketitle
74:
75: \section{Introduction}
76:
77: Brown Dwarfs (BDs) are approximately as abundant as solar mass stars
78: \citep[e.g.][]{Bejar+2001,Chabrier2002}.
79: The typical Jeans' mass in star--forming clouds is of order a solar
80: mass, approximately two orders of magnitude more massive than a BD.
81: This is usually considered the fundamental reason why solar mass
82: stars are so common, but the existence of BDs is a challenge to
83: this simple explanation. The origin of BDs is apparently an important
84: test for the theory of star formation.
85:
86: From a different perspective, we could instead assume that the
87: origin of BDs is quite different from that of more massive stars,
88: in which case their existence would not necessarily provide
89: any constraints to the theory of star formation. This approach is
90: exemplified in the recent suggestion by \cite{Reipurth+Clarke2001}
91: that BDs are the result of the gravitational fragmentation of
92: protostellar disks, but that work does not draw a definite
93: line separating BDs from more massive stars. At what mass does
94: star formation switch from the standard mechanism to the disk
95: fragmentation mode? Most likely not exactly at the hydrogen
96: burning limit of 0.075~m$_{\odot}$, as this limit is defined by
97: nuclear physics and can hardly affect the fragmentation process.
98:
99: There is also no indication of a discontinuity in the stellar
100: initial mass function (IMF) at approximately 0.075~m$_{\odot}$,
101: and no evidence of special properties of young BDs relative to
102: more massive stars, suggesting that perhaps a unique
103: process can explain the formation of both hydrogen burning stars
104: and BDs \citep[e.g.][]{Jayawardhana+2003a,Jayawardhana+2003}.
105:
106: The possibility that BDs and more massive stars have a common origin
107: due to the process of turbulent fragmentation has been recently
108: proposed by \cite{Padoan+Nordlund04bd}. In that work, we
109: show that BDs are predicted by our model for the origin of the
110: stellar IMF from turbulent fragmentation \citep{Padoan+Nordlund02IMF},
111: roughly as frequently as inferred from observations.
112: Here we briefly summarize our theoretical predictions and present
113: new numerical experiments.
114:
115:
116: \section{Stellar and BD Masses from Turbulent Fragmentation}
117:
118: The gas density and velocity fields in star--forming clouds are highly
119: non--linear due to the presence of supersonic turbulence. The kinetic
120: energy of the turbulence is typically 100 times larger than the gas thermal
121: energy on the scale of a few pc (the typical rms Mach number is of
122: order 10) and the gas is roughly isothermal, so that very large
123: compressions due to a complex network of interacting shocks cannot be
124: avoided. Under such conditions the concept of gravitational instability,
125: based on a comparison between gravitational and thermal energies alone
126: in a system with mild perturbations, does not apply. Turbulent density
127: peaks of any size can be formed in the turbulent flow, independent of
128: the Jeans' mass. Peaks that are massive and dense enough (more massive
129: their own Jeans' mass) collapse into protostars, while smaller subcritical
130: ones re-expand into the turbulent flow. This is a process that we call
131: {\it turbulent fragmentation}\footnote{This term was probably first introduced
132: by \cite{Kolesnik+Ogulchanskii90}}, to stress the point that stars and BDs,
133: formed in supersonically turbulent clouds, are not primarily the
134: result of gravitational fragmentation.
135:
136: %
137: \begin{figure*}[ht]
138: \centerline{
139: \epsfxsize=6.5cm \epsfbox{f1a.eps}
140: \epsfxsize=6.5cm \epsfbox{f1b.eps}
141: }
142: \caption[]{Left panel: Analytical mass distributions computed
143: for $\langle n\rangle=10^4$~cm$^{-3}$, $T=10$~K and
144: for three values of the sonic rms Mach number, $M_{\rm S}=5$, 10 and 20
145: (solid lines). The dotted lines show the mass distribution for
146: $T=10$~K, $M_{\rm S}=10$ and $\langle n\rangle=5\times10^3$~cm$^{-3}$
147: (lower plot) and $\langle n\rangle=2\times10^4$~cm$^{-3}$ (upper plot).
148: Right panel: IMF of the cluster IC 348 in Perseus obtained by
149: \cite{Luhman+2003} (solid line histogram) and theoretical IMF computed for
150: $\langle n\rangle=5\times 10^4$~cm$^{-3}$, $T=10$~K and
151: $M_{\rm S}=7$ (dashed line).}
152: \label{f1}
153: \end{figure*}
154: %
155:
156: Because density peaks formed by the turbulent flow need to be larger
157: than their critical mass to collapse, a necessary condition for
158: the formation of BDs by supersonic turbulence is the existence of
159: a finite mass fraction, in the turbulent flow, with density at
160: least as high as the critical one for the collapse of a BD mass
161: density peak. As the PDF of gas density in supersonic turbulence
162: is fully determined by the mean density and the rms Mach number of
163: the turbulent flow \citep{Nordlund+Padoan_Puebla98,Ostriker+99},
164: we can compute this necessary condition for BD formation as an integral
165: over the PDF of gas density. The result, discussed in \cite{Padoan+Nordlund04bd},
166: is that in typical molecular clouds approximately 1\% of the total
167: mass is available for the formation of BDs. This fraction is approximately
168: 10\% for conditions found in regions of cluster formation.
169: Based on this result, the observed frequency of BDs is consistent
170: with the distribution of turbulent pressure, suggesting turbulence
171: may play an important role in the the origin of BDs.
172:
173: In \cite{Padoan+Nordlund02IMF} the stellar IMF is interpreted
174: as the result of the turbulent fragmentation of star--forming
175: clouds. It is shown that: i) The power law slope, $s$, of the
176: stellar IMF is directly related to the turbulent power spectrum
177: slope, $\beta$, $s=3/(4-\beta)$; ii) the abundance of stars of
178: mass smaller than the IMF peak depends on the probability density
179: function (PDF) of the gas density in the turbulent flow.
180: As a result, the IMF is a power law with slope close to Salpeter's
181: value \citep{Salpeter55} at large masses, and the abundance of
182: low mass stars and BDs is a function of the turbulent rms
183: velocity, the mean gas density and the mean temperature.
184: The IMF is fully determined once these physical parameters
185: of the star--forming region are known.
186: As shown in Figure~\ref{f1}, the predicted BD abundance increases with
187: increasing Mach number and density (left panel) and is consistent
188: with the abundance inferred in stellar clusters (right panel).
189:
190: %
191: \begin{figure*}[ht]
192: \centerline{
193: \epsfxsize=13.3cm \epsfbox{f2.eps}
194: }
195: \caption[]{Logarithm of the projected density field from an AMR
196: simulation of isothermal, supersonic hydrodynamic turbulence with
197: an effective resolution of $1024^3$ computational zones (Kritsuk,
198: Padoan \& Norman, in preparation).}
199: \label{f2}
200: \end{figure*}
201: %
202:
203:
204:
205: \section{AMR Experiments}
206:
207:
208: We employ a 3-D parallel structured adaptive mesh refinement (AMR)
209: code, {\em Enzo},\footnote{See {\tt http://cosmos.ucsd.edu/enzo/}}
210: developed at the Laboratory for Computational Astrophysics by Bryan,
211: Norman and collaborators \citep{bryan.99,oshea......04}.
212: {\em Enzo} is a public domain Eulerian grid-based hybrid code which
213: includes hydrodynamic and N-body solvers and uses the AMR algorithm
214: of Berger \& Colella \citep{berger.89} to improve spatial resolution
215: in regions where that is required. The central idea behind AMR is
216: to solve the conservation laws on a grid, adding finer meshes in regions
217: that require enhanced resolution. AMR is spatially-- and time--adaptive,
218: can be used with accurate methods for solving the (magneto)hydrodynamic
219: equations. Mesh refinement can be automatically advanced to an arbitrary
220: level, based on any combination of problem--specific local refinement
221: criteria (e.g., density threshold, strong shocks, high
222: vorticity, large gradients, Jeans length \citep{truelove.....97},
223: cooling time, etc.).
224:
225: With this code, we have run the first AMR turbulence experiment
226: ever attempted, achieving an effective resolution of $1024^3$
227: computational zones (Kritsuk et al. in preparation).
228: The rms Mach number of the turbulent flow is $M_{\rm s}\approx 6$,
229: and we have adopted an isothermal equation of state, periodic
230: boundary conditions, large scale random forcing and mesh refinement
231: based on the local shock or shear amplitude. The logarithm
232: of the projected density field from a snapshot of this simulation
233: is shown in Figure~\ref{f2}. The density field of this simulation
234: shows an incredibly rich structure, with sharp filaments on all
235: scales, broken into even denser peaks. Some of these density
236: peaks are gravitationally unstable and would collapse if gravity
237: were included in the experiment. We refer to these density peaks as
238: {\it turbulent seeds}, as they are the turbulent progenitors of
239: protostars. Turbulent seeds of BD mass are commonly found in
240: the turbulent flow.
241:
242: %
243: \begin{figure*}[ht]
244: \centerline{
245: \epsfxsize=13.7cm \epsfbox{f3.eps}
246: }
247: \caption[]{Left panel: Logarithm of the projected gas density in
248: our AMR simulation of star formation in a turbulent cloud (see text).
249: Right panel: Higher resolution view of a small region inside the
250: dense filament marked by the small square in the left panel. The
251: small square is drawn larger than its actual size, as the magnification
252: factor between the right and the left panels is 840.}
253: \label{f3}
254: \end{figure*}
255: %
256:
257: In order to verify the outcome of the gravitational collapse of the
258: turbulent seeds, we run a second experiment with the same code, where
259: self--gravity is included. In this experiment we use the AMR method
260: to resolve the gravitational collapse. The computational box represents
261: a star--forming cloud of size 5~pc and average number density
262: $500$~cm$^{-3}$. In the collapsing regions we are able to fully
263: resolve protostellar disks down to a scale of approximately 1~AU.
264: This simulation shows that the densest turbulent seeds, found
265: primarily within dense filaments, collapse into protostars surrounded
266: by protostellar disks. Figure~\ref{f3} shows a typical filament
267: containing many collapsing ``seeds'' (left panel). The seeds appear
268: as small dense cores, because of the limited dynamical rage of the
269: image. In fact, visualized with higher levels of refinement, they
270: are seen to be collapsing into protostars with circumstellar disks
271: (right panel of Figure~\ref{f3}). Many of these protostars contain
272: a fraction of a solar mass, some only a BD mass.
273:
274: Further analysis and runs with longer integration time will allow
275: to establish the protostellar mass distribution. However, these
276: simulations already teach us an important lesson: Turbulent seeds of BD
277: mass are common and may collapse into BDs, with no
278: need to form BDs from the fragmentation of protostellar disks.
279: Simulations that do not resolve well the turbulent flow (like perhaps
280: every smooth particle hydrodynamic (SPH) simulation to date) are bound
281: to artificially suppress this important mechanism of BD formation.
282: Furthermore, insufficient numerical resolution may also cause
283: artificial fragmentation of protostellar disks (Fisher et al., in
284: preparation). Recent attempts to support the disk fragmentation origin
285: of BDs, and even deny the importance of turbulent fragmentation,
286: based on SPH simulations \citep{Bate+2002} are affected by both
287: problems. We also warn against a direct application of results from
288: simulations without magnetic fields to present day star formation,
289: because magnetic fields affect the fragmentation mechanism
290: \citep{Padoan+Nordlund99MHD} and the angular momentum transport.
291:
292:
293: \section{Conclusions}
294: High resolution numerical experiments of supersonic
295: turbulence and of star formation in turbulent clouds support the
296: suggestion that BDs are formed primarily from the process of turbulent
297: fragmentation, as {\it turbulent seeds}, like hydrogen burning stars.
298: As frequently pointed out at this workshop, all the observational
299: evidence points to a common origin of stars and BDs. Furthermore,
300: observations have shown that the mass distribution of
301: prestellar condensations is indistinguishable from the stellar IMF
302: \citep{Motte+98,Testi+Sargent98,Onishi+99,Johnstone+2000a,Johnstone+2000b,%
303: Motte+2001,Johnstone+2001,Onishi+2002}, both in the functional shape and in the
304: range of masses, including BD mass cores \citep{Walsh+04}. This direct
305: evidence of the turbulent seeds provides strong support to the
306: turbulent fragmentation origin of BDs.
307:
308: \begin{acknowledgements}
309: The numerical simulations were done using the IBM Data Star system
310: at the San Diego Supercomputer Center with support from NRAC award
311: MCA098020S.
312:
313: \end{acknowledgements}
314:
315: %\bibliographystyle{apj}
316: %\bibliography{apj-jour,padoan,MC,bd,enzo_tech2004}
317:
318: \begin{thebibliography}{28}
319: \expandafter\ifx\csname natexlab\endcsname\relax\def\natexlab#1{#1}\fi
320:
321: \bibitem[{{B{\' e}jar} {et~al.}(2001){B{\' e}jar}, {Mart{\' i}n}, {Zapatero
322: Osorio}, {Rebolo}, {Barrado y Navascu{\' e}s}, {Bailer-Jones}, {Mundt},
323: {Baraffe}, {Chabrier}, \& {Allard}}]{Bejar+2001}
324: {B{\' e}jar}, V.~J.~S., {Mart{\' i}n}, E.~L., {Zapatero Osorio}, M.~R.,
325: {Rebolo}, R., {Barrado y Navascu{\' e}s}, D., {Bailer-Jones}, C.~A.~L.,
326: {Mundt}, R., {Baraffe}, I., {Chabrier}, C., \& {Allard}, F. 2001, \apj, 556,
327: 830
328:
329: \bibitem[{{Bate} {et~al.}(2002){Bate}, {Bonnell}, \& {Bromm}}]{Bate+2002}
330: {Bate}, M.~R., {Bonnell}, I.~A., \& {Bromm}, V. 2002, MNRAS, 332, L65
331:
332: \bibitem[{Berger \& Colella(1989)}]{berger.89}
333: Berger, M.~J. \& Colella, P. 1989, \jcp, 82, 64
334:
335: \bibitem[{Bryan \& Norman(1999)}]{bryan.99}
336: Bryan, G.~L. \& Norman, M.~L. 1999, in IMA, Vol. 117, Adaptive Mesh Refinement
337: (SAMR) Grid Methods, ed. S.~B. Baden, N.~P. Chrisochoides, D.~Gannon, \&
338: M.~L. Norman, New York: Springer, 165
339:
340: \bibitem[{{Chabrier}(2002)}]{Chabrier2002}
341: {Chabrier}, G. 2002, \apj, 567, 304
342:
343: \bibitem[{{Jayawardhana} {et~al.}(2003{\natexlab{b}}){Jayawardhana}, {Ardila},
344: {Stelzer}, \& {Haisch}}]{Jayawardhana+2003}
345: {Jayawardhana}, R., {Ardila}, D.~R., {Stelzer}, B., \& {Haisch}, K.~E.
346: 2003{\natexlab{a}}, \aj, 126, 1515
347:
348: \bibitem[{{Jayawardhana} {et~al.}(2003{\natexlab{a}}){Jayawardhana}, {Mohanty},
349: \& {Basri}}]{Jayawardhana+2003a}
350: {Jayawardhana}, R., {Mohanty}, S., \& {Basri}, G. 2003{\natexlab{b}}, \apj,
351: 592, 282
352:
353: \bibitem[{{Johnstone} {et~al.}(2001){Johnstone}, {Fich}, {Mitchell}, \&
354: {Moriarty-Schieven}}]{Johnstone+2001}
355: {Johnstone}, D., {Fich}, M., {Mitchell}, G.~F., \& {Moriarty-Schieven}, G.
356: 2001, \apj, 559, 307
357:
358: \bibitem[{{Johnstone} {et~al.}(2000{\natexlab{a}}){Johnstone}, {Wilson},
359: {Moriarty-Schieven}, {Giannakopoulou-Creighton}, \&
360: {Gregersen}}]{Johnstone+2000a}
361: {Johnstone}, D., {Wilson}, C.~D., {Moriarty-Schieven}, G.,
362: {Giannakopoulou-Creighton}, J., \& {Gregersen}, E. 2000{\natexlab{a}}, \apjs,
363: 131, 505
364:
365: \bibitem[{{Johnstone} {et~al.}(2000{\natexlab{b}}){Johnstone}, {Wilson},
366: {Moriarty-Schieven}, {Joncas}, {Smith}, {Gregersen}, \&
367: {Fich}}]{Johnstone+2000b}
368: {Johnstone}, D., {Wilson}, C.~D., {Moriarty-Schieven}, G., {Joncas}, G.,
369: {Smith}, G., {Gregersen}, E., \& {Fich}, M. 2000{\natexlab{b}}, \apj, 545,
370: 327
371:
372: \bibitem[{{Kolesnik} \& {Ogul'Chanskii}(1990)}]{Kolesnik+Ogulchanskii90}
373: {Kolesnik}, I. \& {Ogul'Chanskii}, I.~I. 1990, in ASSL Vol. 162: Physical
374: Processes in Fragmentation and Star Formation, 81--86
375:
376: \bibitem[{{Luhman} {et~al.}(2003){Luhman}, {Stauffer}, {Muench}, {Rieke},
377: {Lada}, {Bouvier}, \& {Lada}}]{Luhman+2003}
378: {Luhman}, K.~L., {Stauffer}, J.~R., {Muench}, A.~A., {Rieke}, G.~H., {Lada},
379: E.~A., {Bouvier}, J., \& {Lada}, C.~J. 2003, \apj, 593, 1093
380:
381: \bibitem[{{Motte} {et~al.}(2001){Motte}, {Andr{\' e}}, {Ward-Thompson}, \&
382: {Bontemps}}]{Motte+2001}
383: {Motte}, F., {Andr{\' e}}, P., {Ward-Thompson}, D., \& {Bontemps}, S. 2001,
384: \aap, 372, L41
385:
386: \bibitem[{{Motte} {et~al.}(1998){Motte}, {Andr\'{e}}, \& {Neri}}]{Motte+98}
387: {Motte}, F., {Andr\'{e}}, P., \& {Neri}, R. 1998, A\&A, 336, 150
388:
389: \bibitem[{Nordlund \& Padoan(1999)}]{Nordlund+Padoan_Puebla98}
390: Nordlund, {\AA}. \& Padoan, P. 1999, in Interstellar Turbulence, ed. J.~Franco
391: \& A.~Carrami{\~n}ana (Cambridge University Press), 218
392:
393: \bibitem[{{Onishi} {et~al.}(1999){Onishi}, {Mizuno}, {Kawamura}, \&
394: {Fukui}}]{Onishi+99}
395: {Onishi}, T., {Mizuno}, A., {Kawamura}, A., \& {Fukui}, Y. 1999, in Star
396: Formation 1999, Proceedings of Star Formation 1999, held in Nagoya, Japan,
397: June 21 - 25, 1999, Editor: T. Nakamoto, Nobeyama Radio Observatory, p.
398: 153-158, 153--158
399:
400: \bibitem[{{Onishi} {et~al.}(2002){Onishi}, {Mizuno}, {Kawamura}, {Tachihara},
401: \& {Fukui}}]{Onishi+2002}
402: {Onishi}, T., {Mizuno}, A., {Kawamura}, A., {Tachihara}, K., \& {Fukui}, Y.
403: 2002, \apj, 575, 950
404:
405: \bibitem[{{O'Shea} {et~al.}(2004){O'Shea}, {Bryan}, {Bordner}, {Norman},
406: {Abel}, {Harkness}, \& {Kritsuk}}]{oshea......04}
407: {O'Shea}, B.~W., {Bryan}, G., {Bordner}, J., {Norman}, M.~L., {Abel}, T.,
408: {Harkness}, R., \& {Kritsuk}, A. 2004, astro-ph/0403044
409:
410: \bibitem[{{Ostriker} {et~al.}(1999){Ostriker}, {Gammie}, \&
411: {Stone}}]{Ostriker+99}
412: {Ostriker}, E.~C., {Gammie}, C.~F., \& {Stone}, J.~M. 1999, ApJ, 513, 259
413:
414: \bibitem[{Padoan \& Nordlund(1999)}]{Padoan+Nordlund99MHD}
415: Padoan, P. \& Nordlund, {\AA}. 1999, ApJ, 526, 279
416:
417: \bibitem[{{Padoan} \& {Nordlund}(2002)}]{Padoan+Nordlund02IMF}
418: {Padoan}, P. \& {Nordlund}, {\AA}. 2002, \apj, 576, 870
419:
420: \bibitem[{{Padoan} \& {Nordlund}(2004)}]{Padoan+Nordlund04bd}
421: ---. 2004, ApJ, 617
422:
423: \bibitem[{{Reipurth} \& {Clarke}(2001)}]{Reipurth+Clarke2001}
424: {Reipurth}, B. \& {Clarke}, C. 2001, \aj, 122, 432
425:
426: \bibitem[{{Salpeter}(1955)}]{Salpeter55}
427: {Salpeter}, E.~E. 1955, \apj, 121, 161
428:
429: \bibitem[{{Testi} \& {Sargent}(1998)}]{Testi+Sargent98}
430: {Testi}, L. \& {Sargent}, A.~I. 1998, ApJL, 508, L91
431:
432: \bibitem[{{Truelove} {et~al.}(1997){Truelove}, {Klein}, {McKee}, {Holliman},
433: {Howell}, \& {Greenough}}]{truelove.....97}
434: {Truelove}, J.~K., {Klein}, R.~I., {McKee}, C.~F., {Holliman}, J.~H., {Howell},
435: L.~H., \& {Greenough}, J.~A. 1997, \apjl, 489, L179
436:
437: \bibitem[{Walsh {et~al.}(2004)Walsh, Di~Francesco, Myers, Bourke, \&
438: Wilner}]{Walsh+04}
439: Walsh, A.~J., Di~Francesco, J., Myers, P.~C., Bourke, T.~L., \& Wilner, D.~J.
440: 2004, "93 N2H+ Clumps in NGC 1333", poster from "Cores, Disks, Jets and
441: Outflows in Low and High Mass Star Forming Environments: Observations, Theory
442: and Simulations", Banff, Alberta, Canada, July 12-16, 2004
443:
444: \end{thebibliography}
445:
446:
447: \end{document}
448: